Interactive Fluidic State Machines for Soft Robotics

Project Information

ILUMIS

Grant agreement ID: 101076036

DOI

10.3030/101076036

EC signature date 17 January 2023

Start date 1 May 2023

End date 30 April 2028

Funded under

• European Research Council (ERC)

Total cost

€ 1 497 000,00

EU contribution

€ 1 497 000,00

Coordinated by

KATHOLIEKE UNIVERSITEIT LEUVEN
Belgium

Periodic Reporting for period 1 - ILUMIS (Interactive Fluidic State Machines for Soft Robotics)

Reporting period: 2023-05-01 to 2025-10-31

Summary of the context and overall objectives of the project

Actuation, energy storage, sensing, and logic are four functionalities of both natural and artificial organisms, giving them the ability to thrive in their environment. The blueprint of conventional robots localizes these functionalities in discrete components supported by rigid materials. However, in soft robots that consist of compliant materials, localization of functionality severely limits autonomous operation and intelligent behavior. This limitation is the result of the functional architecture, not of the used materials. Alternatively and as demonstrated in nature by the common octopus, the distribution of these four functionalities throughout the body allows to overcome these limitations. This concept of ‘functional embodiment’ is currently non-existing in soft robotics. Project ILUMIS creates soft robots with embodied functionality by transitioning from a conventional robotic architecture to a fluidic network architecture. Further, by incorporating nonlinearities in all the network elements, the global system acts as a state machine, meaning that the output not only depends on the input, but also on its internal state. How to navigate this state space is encoded within the nonlinearities, creating embodied logic. Energy and actuation are embodied and intricately linked to the elastic deformations of the components in the network, powering the actions of the soft robot. By creating network components that are sensitive to triggers from the environment, embodied sensing emerges, leading to truly interactive fluidic state machines. ILUMIS surmounts the main challenges of inverse design, where a desired behavior requires the optimization of a network of nonlinear structures. Thereby ILUMIS creates a new blueprint for soft robotic design with embodied functionality that closes the gap with nature’s soft organisms.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

The research in project ILUMIS is organized in three fundamental pillars, that each study a particular nonlinear element in the fluidic network, being inflatable structures, interconnections and fluids. These elements are then combined into fluidic networks, constituting pillar four. The activities and achievements are discussed for each pillar below.

Inflatable structures: The work concentrated on (i) developing new types of inflatable actuators that display large circumferential strains, as published in [Advanced Materials Technologies, 9(4), p.2301662]; (ii) creating a design methodology for highly nonlinear disk spring actuators and a framework for sequencing their actuation, as published in [Advanced Materials, 35(35), 2301487]; (iii) devising inflatable metamaterial sleeve actuators that break motion symmetry with one fluidic input, as published in [Advanced Intelligent Systems, 2500157]. More recent work focuses on the analysis of parallel connections of inflatable structures, as presented at APS March meeting 2025 (oral presentation “Interacting Fluidic Hysterons”).

Interconnections: We developed a reconfigurable valve that is used in an oscillatory circuit, as published in [Advanced Science, 11(43), 2470264.]. More recent work focuses on crating a highly nonlinear valve, inspired by biological neurons.

Fluids: We created a new type of fluid that we coined metafluids, and essentially consist of submerged hollow shells in a liquid. The results were published in [Nature, 628(8008), 545-550]. More recent work focuses on the design of shells that are more nonlinear, as presented at APS March meeting 2025 (oral presentation “Multi-step Pathways of a Spherical Shell Structure with a Hierarchical Architecture”).

Fluidic Networks: we studied how fluidic oscillations can occur from only hysteresis in the pneumatic domain, coupled with negative feedback. The results were published [Advanced Intelligent Systems, 2400695]

Overarching these pillars, we also published papers that (i) analyze how soft materials are affected by the environment [Polymers, 15(13), 2964]; (ii) disseminate a general vision on physical control [Science Robotics 10 (102), eadw7660]. Our methodology on analyzing networks was also presented at APS March meeting 2025 (oral presentation “On the coupling of non-linear inflatables and springs: Island-hopping”).

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

Again, the same 4 pillars will be used to describe the results beyond the state of the arts.

Inflatable structures: We have developed a methodology for the design of actuation sequences that is universal for all active systems that snap, it is usable for a wide range of domains. The actual snapping component is now a standard part that is regularly used in soft robotics literature.

Interconnections: Further research is needed to create a standard part that can be taken up by the scientific community. This is currently an active line of research.

Fluids: Further research is needed that increases the manufacturability of the metafluids, directly incorporating gasses in the hollow shells. This is currently an active line of research where we are also looking into advanced demonstrators that can lead to further exploitation.

Fluidic Networks: initial insights are gathered to formalize a framework for the design of oscillatory circuits, that can instigate complex actuation patterns. This is currently an active line of research.

New blueprint for soft robots that embody actuation, energy storage, sensing, and logic.