Digital Theses Archive


Tesi etd-04222017-104829

Type of thesis
Study and Implementation of Algorithms Inspired by Movements and Control Strategies of Plant Roots
Scientific disciplinary sector
INGEGNERIA - Biorobotics
  • Behavior analysis
  • Behavior-based control
  • Bioinspiration
  • Plant-inspired robot
  • Robot control
Exam session start date
Aim of the present work is to implement innovative solutions for the<br>movements and control of plant-inspired robots, by extracting fundamental<br>behavioral rules from plant roots.<br>Biological systems have been source of inspiration for engineers and roboticists<br>since long time; especially animals, or even simpler organisms such as bacteria,<br>have inspired morphology and behavior of robots and optimization algorithms.<br>Plants have instead been taken as models in robotics only relatively recently. Being<br>sessile organisms and devoid of a neural system, they have rarely sparked interest<br>in robotics community. On the contrary, plants show a numerous series of<br>movements and communication abilities that can offer new cues for designing<br>innovative robots for exploration tasks.<br>In particular, root apparatus and its apexes represent the organ delegates to<br>anchor and forage the whole plant system and has to mediate among many stimuli<br>and needs. These observations motivated the present dissertation which presents a<br>series of data from literature and purposely performed experiments to extract<br>specifications and design rules from plant roots for the development of innovative<br>penetration strategies and control algorithms.<br>Results of a deep study on a peculiar movement observed in plant roots, i.e.,<br>circumnutation, demonstrate its role in optimizing root soil penetration. Such<br>strategy can indeed be considered relevant for both real plants and artificial<br>penetration devices, helping to employ less forces and energy in digging tasks.<br>Moreover, the same movement has been investigated as mechanism in plant roots<br>for resources exploration and exploitation; by comparing it with the oscillatory<br>movement actuated during chemical stimulation by a relatively similar simple<br>organism, i.e., Drosophila larvae.<br>Two different control strategies are also presented, which take inspiration from<br>tropic responses and from the uptake-kinetic of nutrients by roots. In particular,<br>the latter, because of a distributed knowledge achieved by internal communication<br>channels, reveals an interesting emerging collaborative behavior among roots for the<br>optimization of plant wellness.<br>All the investigated strategies are implemented for the control of a plant-inspired<br>robot, called Plantoid, which nicely validates the hypothesis on circumnutations’<br>role and collaborative behavior done on the biological counterpart.<br>The expected impacts of present study are twofold: I) to demonstrate that plant<br>roots are a valid model for the realization of innovative technological solutions,<br>specifically control strategies, potentially useful to expand algorithms in the swarm<br>intelligence field; II) to provide an effective approach for the validation of biological<br>hypotheses formulated on plant movements and behavior.