Short term visitor's project
Bio-inspiration, i.e. using phenomena in biology to make innovation in science and technology, as a general concept is all but new. Yet, its application to robotics is massively innovating the field, re-thinking e.g. patterns of movement or growth, or sensing or actuation abilities. The bioinspired evolution of robotics also supports the effectiveness of soft bodies, as living organisms exploit soft tissues and compliant structures to move in complex environments. This new class of robots is expected to act in unstructured scenarios (e.g., locomotion in un-certain terrains, manipulation of unknown objects, accomplishment of non-predetermined tasks, etc.) and interact more safely with humans. The complexity in developing bioinspired soft robots is increased by the need of mimicking biological system capabilities in being energetically efficient, in changing their morphology, in adapting their body and functionality in their lifetime in a continuously changing environment. Hence, the challenge ahead for soft robotics is to further develop the abilities for robots to grow, evolve, self-heal, develop, and bio-degrade, which are the ways that robots can adapt their morphology to the environment.
Only recently, plants have been considered as a model in robotics. The first attempt in robotics to imitate plant roots was proposed by Barbara Mazzolai and her group. Roots are the organs delegated to the foraging and anchoring of plants. While performing these tasks, roots need to adapt to the environment, avoid obstacles, penetrate soils at different impedance, and follow nutrient and water gradients. The robots inspired by these capabilities are named Plantoids and they represented the first time worldwide that robots were inspired by plant roots, equipped with distributed sensing, actuation, and behaviours for soil exploration.
The current version of the root-like robot grows through a monotonic process that continuously adds new material, integrating inside its robotic head an additive manufacturing technique based on Fusion Deposition Modelling for the layer-by-layer deposition of a thermoplastic filament. However, although plantoid robots represent a potential revolution in soil exploration, the complexity of the medium in terms of high pressure and friction requires an optimization of the design of the robotic roots, with abilities of morphological adaptation and penetration strategies more similar to the natural counterpart.
The objective and main novelty of the project are to reconstruct the in vivo biomechanics of roots in soil over time (4D biomechanics), with the characterisation of the phases of growth, elongation, circumnutation, by using x-ray tomography. The development of mathematical models of soil and robotic solutions will support the measurement and reconstruction of the different phases of root growth and the understanding of how the natural system adapts to different external environmental conditions.
New design rules of growing robots for less invasive and destructive and energy-efficient soil penetrations are the main scientific and technological expected outcomes of this project. These robotic autonomous systems are envisioned to find applications in many different scenarios: for civil engineering, for environmental monitoring, for efficient and sustainable agriculture, or for space explorations.
- PI: Barbara Mazzolai (visitor)
- Co-PI: Luc Sibille