Plantoid

A plantoid is a plant-inspired robot. Plantoids belong to the field of biorobotics where biological knowledge is used to develop innovative methodologies and technologies and bio-inspired robots serve as a tool to study living organisms. Characteristics and mechanisms of interacting with the environment such as the sensing and growing features of plant roots and the grasping behavior of tendrils have inspired plantoid designs. Potential applications proposed for plantoids include soil drilling, environmental monitoring, orthopedic supports and surgical robots.

It has been recognized that plant actuation and morphing mechanisms have applications in soft robotics. Plant cells and tissues provide movement capability and structural rigidity without the distinction between ‘actuators’ and ‘structures’ common to animal systems. Compared to animal muscular motions, plant movements are water driven, distributed, energy efficient and integrative. Inspiration from plants provides an opportunity to design movement with less part numbers. Plants can also provide inspiration for building soft robots out of compliant materials with the ability to interact in a soft and safe manner with the environment and humans.

The concept of a plantoid was proposed by Mazzolai et al. (2010) to consist of task-specialized modules representing their natural counterparts and functionalities. The authors, lead by Stefano Mancuso, divide a plantoid can into three main sub-systems: 1) a main body with batteries, electronics and radio systems; 2) the root system, with electro-osmotic actuators; 3) the root apex with sensors. Furthermore, leaves are proposed to include photovoltaic cells.

The reiteration and swarm behavior of roots are aspects that Mazzolai et al. (2010) suggest are relevant to robotics. The growth of roots appears coordinated despite a central nervous system. The authors make an analogy between the ability of plant roots to navigate towards resources such as water and nutrients and the collective behavior in birds during long-range migration. The plantoid concept proposed by Mazzolai et al. is of several distributed and self-organizing modules, rather than a centralized system.

The European Commission funded The Plantoid Project from 2012 to 2015. The goals of the project were 1) to abstract and synthesize with robotic artefacts the principles that enable plant roots to explore and adapt to underground environments and 2) formulate testable hypotheses and models.

The Plantoid Project was partly inspired by plant intelligence research findings from researchers such as Stefano Mancuso, Elisabeth Von Volkenburgh, Monica Gagliano and Frantisek Baluska. Baluska’s research into root behavior and responding to gravity and light demonstrated that plants make decisions based on sensory perceptions. Plants have also been found to communicate with other plants nearby by releasing volatiles as an alert to pathogen attacks. Mancuso was a partner in the Plantoid Project and the coordinator was Barbara Mazzolai at the Center for Micro-BioRobotics, Istituto Italiano di Tecnologia (IIT).

Principles, theoretical investigations and experimental measurements towards designing a mechatronic system called apex, for soil exploration, inspired by plant roots was described in 2011. The research team included Barbara Mazzolai and was lead by Paolo Dario at the Center for Research in Microengineering and Advanced Robotics Technologies and Systems Laboratories, Scuola Superiore Sant’Anna, Pontedera, Italy. Taking inspiration from the way a root moves through substrate, orienting with respect to gravity and locating water and nutrients, the mechatronic apex is embedded with a gravity sensor, a soil moisture gradient detector and the electronics for sensory data acquisition and steering. A prototype of the mechatronic apex system was tested with a bio-inspired algorithm reproduced the gravitropism and hydrotropism behaviors of plants. An osmotic actuator module was also presented.

Stretching processes and mechanoperception of the plant root inspired the development of a soft robotics approach where capacitive sensitive elements conform to the shape of the soft body and follow its deformations. Plant cells possess mechanoperception in the ability of their cells to deform in response to external and internal mechanical forces. Plant roots respond to touch and bending with cell shape change and an internal signal of increased concentration of cytosolic calcium, Ca2+. The concept and design of a robotic soft body based on these plant root characteristics and built out of elastomers and conductive textiles was presented in Scientific Reports in 2015 by a team led by Lucia Beccai at the Istituto Italiano di Tecnologia (ITT). The concave and convex sides of a bent body of distinguishable by their sensory responses. The sensing elements and the body act as one single entity rather than separate integrated parts.

The design and prototype of a root-like growing robot was published in PLoS ONE in 2014 with a research team lead by Barbara Mazzolai at Center for Micro-BioRobotics at the Istituto Italiano di Tecnologia (IIT). Their device is designed to penetrate soil and develops its own structure using an additive layering technique where each layer of new material is added to the tip similar to in plant roots where new cells are added at the root apex by mitosis. The deposition produces both a motive force at the tip and a hollow tubular structure extending to the surface of the soil and anchored to the soil. By omitting peripheral friction, soil penetration uses less energy than pushing the soil from the base of the penetration system. A path for delivering materials is provided by the tubular structure.

Design of root-like soil penetration device from Figure 3 of Sadeghi A. et al. PLoS ONE 9(2): e90139.

Potential applications for such a system include autonomous tunneling or space applications. As an approach to assembly, the growth process could be used for construction of structures. The root-like growing robot could also serve as a model to validate hypotheses on plant growth.

In 2017, Mazzolai’s team presented another root-like growing design that included a mechanism similar to cell sloughing and the ability to grow straight or to bend. . In plant roots cells grow outward from the tips and the sloughing of cells provide a low frictional interaction with the soil. These characteristics were incorporated into their plantoid prototype. Their system has an outward movement of a soft sleeve from the internal hole of a rigid tube to the external face. In addition the positive effect of root hairs in providing anchorage was imitated with lateral hairs which prevented upward movement. The body of the robot does not move with respect to the soil. The apical part performs the penetration using the growing mechanism which is a material deposition process achieved by the integration of a 3D printer inside the root. The robotic roots are able to grow straight or bend depending on the location of the deposition at the tip.

Figure 1 from Sadeghi et al. Soft Robot (2017) 4(3):221-223 showing a general view of a plantoid with a growing root mechanism.

Each robot root includes a tubular body, a growing head and a tip with a sensor that commands the robot behaviors. Layer-by-layer deposition of the fused material in a tubular shape for the root body is achieved by simultaneous actions of feeding and rotational plotting. The 3D printer deposits polylactic acid (PLA) filament. By controlling the heater temperature and feeding speed the compliancy and softness of fused material can be modified. The plantoid with growing plant-like roots has potential applications in soil drilling, soil monitoring. Other suggested applications include passing oxygen, drugs or food in rescue scenarios and also in medical applications, such a device could be used for moving surgical tools.

In 2019 a team led by Barbara Mazzolai at the Center for Micro-BioRobotics at the Istituto Italiano di Tecnologia (IIT) developed the first soft robot that mimics the behavior of tendrils. The ITT group’s plantoid’s reversible osmotic actuation strategy is inspired by the way plants reversibly modify intracellular turgor to tune stiffness and achieve macroscopic movements. Plant cell turgor (pressure) is set up using water influx due to osmolyte concentration gradients through the natural osmotic barrier established by the cell wall and plasma membrane. In plants the cytosolic osmolyte system contains proteins and small molecules in an aqueous gel solution. Plant cells control osmolyte gradients to generate reversible movements.

Inspired by grapevines, the planetoid is made of plastic, polyethylene terephthalate (PET), and is able to curl around and climb a support to secure position, similar to vines. The artificial tendril works by the same physical process of water transportation in plants. A polysulfone tube of electrically charged particles in liquid at the bottom of the robot acts as an osmotic membrane. The tube snakes in between layers of carbon fiber fabrics which function as electrodes so that when the unit is connected to a 1.3 volt battery the ions are attracted to, move and attach to the surface of the flexible cloth. The movement of the charged particles causes the liquid to flow, which makes the tendril move in a coiling motion. When the battery is detached an opposite movement occurs. Potential applications for this plantoid include wearable, flexible orthopedic supports that adapt to a patient’s needs and tendrils with sensors or cameras that can monitor pollution or rescue people.

Figure 1 from Must et al. Nature Communications 10, Article number: 344 (2019). Osmosis-based reversible actuation.

A different approach to the development of climbing robots inspired by plant tendrils was presented by a group led by Camilla Pandolfi at the European Space Agency-Advanced Concepts Team in the Netherlands. In their 2015 paper they evaluated the movements and behaviors of the tendril, develop a robotic model and a kinematic simulator was used to visualize the system. Proof of concept prototypes were made of smart materials. Their design was based on climbing by searching for support, grasping-by-coiling and pulling behaviors. Using computer simulation, the searching phase was reproduced using a centralized motion so the tendrils span a cone in a 3D motion. If grasping-coiling sections of the device touch an obstacle, the circumnutation motion stops and the area that touched a surface reacts by bending in the direction of the stimulus. Shape Memory Alloys (SMAs) were chosen as material for their proof-of -concept prototypes, making use the ability of SMAs are able to recover a predetermined shape when heated.

The sensing principles of cucumber tendrils inspired the design of a sleeve for soft surgical manipulators, which could be developed into Robot-assisted Minimally Invasive Surgery (RMIS), where robotic manipulators pass through small incisions into the patient’s body. Cucumber tendrils have tactile papillae on their tendrils, which they use for climbing. A research team led by Kaspar Althoefer at the Centre for Advanced Robotic at Queen Mary University of London presented their design of soft manipulators with miniature sensing elements across the surface in the 2014 IEEE International Conference on Robotics and Automation (ICRA). The sensor network is reported to be capable of acquiring tactile information or haptic feedback for the surgeon. Each sensing element is a retractable hemispherical tactile that measures applied pressure. Optic fibers are used to transfer light signals modulated by the applied pressure from the sensing element to the proximal end of the robot arm. The device also takes inspiration from octopus morphology.

The International Laboratory of Plant Neurobiology (LINV) in Florence, Italy is developing a plantoid that may be used to explore Marian soil by dropping mechanical “pods” able to communicate with a central “stem”, which would send data back to Earth.

A robotic venus flytrap was designed based on the rapid movements of carnivorous plants, triggered by antenna-like sensors. In the venus flytrap, prey moves the trigger hairs which electro-elastically send an electric signal to the internal ions in the lobe to migrate outwardly. This causes the jaw-like lobes close and capture the prey. The manner by which the lobes bend inward is similar to ionic polymeric metal composites (IPMCs) bending in an electric field. The robotic venus flytrap has IPMC lobes with a common electrode in the middle of one end to act like a spine. IPMC due to its mechano-electrical sensing characteristics was used for the trigger hairs as well.

Mazzolai is the coordinator of GrowBot, a European collaborative project that aims to develop climbing plants. The GrowBot objective is to develop low-mass and low-volume robots that can anchor themselves, negotiate voids and climb in places where robots based on wheels, legs or rails would not navigate well. GrowBot is a €6,997,482.50 project funded under the FET Proactive: emerging paradigms and communities Research and Innovation Action Grant agreement n. 824074 which began in 2019.