There is a growing need to develop robots that can interact with humans, our living environment, natural terrain, and afflicted areas. These robotic systems should be safe, resilient, and able to adapt to dynamically changing environments. The traditional robotic approach toward these goals has mainly focused on hard and rigid mechanical system with control systems designed in a centralized manner so as not to damage humans and environments. The problem with this strategy is that it requires huge amount of computational cost and yet cannot deal with an unexpected situation.

    An alternative strategy is to build the robots, mostly or completely, from soft and flexible materials. The mechanical softness gives roboticists new opportunities to design safe, interactive machines and exploit novel sensing and actuation technologies. However, when a machine is built from soft materials, its movements are exceedingly difficult to control with the traditional centralized control scheme because of intrinsic non-linear and elastic responses associated with the deformation. Currently, there is no systematic method for controlling the movements of a system made of soft material components.

    My pivotal research theme is to design mechanically soft robotic systems inspired by soft-bodied animals and to understand how soft materials contribute to the control of robots’ movement. In contrast to traditional hard and rigid robotic systems, bodies of animals are made predominantly of soft materials such as muscles, tendons, and skin that can deform easily in three-dimensional space. Animals—even living systems without a brain, such as amoeba and true slime mold—have the capacity to orchestrate movements with enormously large degrees of freedom and generate adaptive behaviors. In order to understand the underlying mechanisms, my research is inspired by such living systems (i.e., true slime mold and caterpillar) extracting decentralized control principles for highly deformable moveable structures. Some of my research in this area is briefly summarized below.

1 Autonomous decentralized control of Amoeba-like Soft Robot Inspired by True Slime Mold
any simple organisms can coordinate their movements in the absence of a central nervous system. These organisms rely on decentralized control and exploit internal and external mechanical interactions. Such decentralized control is now thought to be an important component of locomotion in many animals of even greater complexity (e.g., central pattern generators in vertebrates, distributed nervous systems in insects, starfish and jellyfish).

    To begin understanding the principles that underlie decentralized control of locomotion, I have focused on the movements of an extraordinary living organism, the plasmodium of true slime mold. The true slime mold is a large amoeba-like unicellular multi-nucleated organism, which exhibits amazingly adaptive behaviors including, taxis and exploration behavior despite its lack of a nervous system. The key component driving its motion are coupled biochemical oscillators distributed throughout the body. These oscillators generate rhythmic mechanical contractions (whose cycle is 1-2 minutes), which in turn induce protoplasmic streaming between the body parts. By changing these patterns and mechanical interaction between the oscillators, the true slime mold generates qualitatively different locomotion/behaviors: a traveling wave for taxis locomotion and a spiral wave for exploratory behavior. In addition to these membrane contractions, the organism can also generate and degenerate pseudopods (temporary protrusions produced by sol-gel transitions of the protoplasm) to adapt to the environment over a longer time-scale. This allows the animal to form complex morphologies for connecting separated food sources or for minimizing its exposure to toxic and hazardous chemicals. These multi-timescale adaptive mechanisms co-exit in the body and interact to enhance survival.

    From an engineering point of view, the true slime mold realizes the following important functions, all of which are difficult to substantiate with a classic centralized controller:
(i) Controlling and taming large degrees of freedom of the soft material(s);
(ii) Generating versatile behaviors and transition/switching between them; and
(iii) Combining two different adaptive mechanisms seamlessly and synergistically to enhance the adaptability.
I have realized these three fundamental functions to build life-like resilient and intelligent robots.

1.1 Controlling and taming large degrees of freedom inside the soft-body [Biol. Cybern. 2010, UC 2011]

The underlying mechanism of taxis locomotion by the true slime mold involves local sensory feedback of the bio-chemical oscillators. This feedback can be extremely simple: when a local oscillator is pushed strongly by the protoplasm under high pressure, it is stretched even if it is trying to contract. Based on this biological finding, I designed a mathematical model of the local sensory feedback, and embedded it into a slime mold robot (both the simulation model and real robot on the YouTube videos bellow) as an autonomous decentralized control system. There are two unique features of this robot:

  • the robot has an outer skin that forms a truly soft and deformable body using passive actuators (that can change its resting length dynamically) and a central balloon (i.e., protoplasm for this robot) that transmits mechanical interaction between the actuators (i.e., the volume constraint of the air inside the balloon); and
  • a fully decentralized control (i.e., coupled oscillators to control the resting length) with local sensory feedback realized by exploiting the mechanical interaction between the actuators.

Simulation and experimental results show that this robot exhibits truly supple and adaptive locomotion without relying on any hierarchical structure. The results obtained indicated that the autonomous decentralized control is a promising approach for soft-bodied robotic systems. This control scheme has also been applied to snake-like and quadruped robots by my colleagues in the Ishiguro laboratory (Owaki et al. Interface 2012, Sato et al. Bioinspir. Biomim. 2011).

A soft-bodied amoeboid robot "Slimy" (simulated result)

A Soft Deformable Amoeboid Robot Inspired by Plasmodium of True Slime Mold -Slimy-

1.2 Generating versatile behaviors and transition/switching between them [Artificial life 2013, Bioinspir. Biomim 2013, Adaptive behavior 2015a]

nother advantage of soft robots is that versatile behaviors can be generated using the large degrees of freedom in their bodies. This is one of the fundamental strategies for animals adapting to unexpected situations. In order to investigate the capability of the proposed controller, I simplified the true slime mold robot into hydrostatically coupled oscillators (consisting of passive actuators, air cylinders, and tubes to connect them). Despite this simplicity, the real physical robot produced versatile oscillatory patterns and spontaneous transitions among the patterns by exploiting the mechanical (hydrostatic) interplay (the simulation model was published in
Artificial life 2013, and the real physical robot and further analysis was published in Bioinspir. Biomim 2013). Based on the oscillator system, we built a modular robot with local stiffness changes that depended on the presence of an attractant; the robot was able to switch from exploratory to taxis locomotion (Adaptive behavior 2015a).

    These results also indicate that mechanical interactions can transfer information between oscillators (distributed controller) without a designing a specific communication process between them. This interaction between local controllers is unique and sharply contrasts with many proposed CPG-based controllers using ‘well-designed networks’. I also believe that studying and reproducing these behaviors of the coupled oscillators can contribute to understanding more universal motion control of animals, such as biped and quadruped locomotion, because coupled oscillator systems and rhythmic motion are ubiquitous in all living systems.

1.3 Combining two different adaptive mechanism seamlessly and synergistically to enhance the adaptively [Adaptive behavior 2015b]

In real living systems adaptive mechanisms with different time constants can co-exist without causing conflicts in the body (e.g., reflex, learning, growth and evolution). This enables living systems to survive in the face of overwhelming environmental changes, which a single adaptation mechanism would not allow. In the true slime mold at least two adaptive mechanisms exist: one is a contraction mechanism that generates cyclic oscillations with a period of 1-2 minutes and another is a morphological change producing and eliminating pseudopods over a time-scale 10 times longer. Inspired by this, we have designed a mathematical model and real physical robot incorporating these two mechanisms as decentralized controllers. Numerical and experimental results show that by combining the controllers with different time constants, a robot can use the proposed model to successfully negotiate a narrow aisle by deforming its body shape dynamically.

Motion control of caterpillar-like soft robots [IROS 2013, ICRA 2014]
Soft animals and robots face a fundamental problem of controlling movements with unlimited degrees of freedom. To examine this problem in more detail I have developed 3-D printed soft robots based on a highly tractable animal model, Manduca sexta. Most animal locomotion involves two important control mechanisms: rhythm generation and postural maintenance. According to this, I have examined how to actuate soft materials with rhythm generation and postural maintenance. Using only two actuators I have shown that rhythmic motor patterns can generate both inching and crawling locomotion, which is similar to the real caterpillar (movie: the left YouTube video). Furthermore, using three actuators I also have demonstrated that imbalanced contraction forces between antagonistically arranged actuators on the robot also allows the robot to steer (movie: the right YouTube video).

Soft-bodied Caterpillar-like Robot

YouTube 動画

3 Real-time Tunable Spring
[IROS ’06]
Another approach to controlling large degree of freedom movements is to better exploit the mechanical properties of the actuators and physical structures. I have developed several types of actuators, based on a `Real-time Tunable Spring.’  One actuator (used in the amoeba-like soft robot) can change its resting length and therefore its stiffness by winding/unwinding the coiled spring. Another can change its stiffness actively without changing its resting length. I developed an elastic ring whose thickness distribution was carefully designed such that the dynamic stiffness can be altered by rotating the elastic ring (the Youtube Video
below).  These passive actuators allow the robot’s body parts to interact through body dynamics (e.g., continuum deformation and body structures) without restraining the overall softness of the robots. This also means that a robot designer can focus on the interaction manner, not designing precise trajectory of the each body part.

Real-time Tunable Spring (elasticity variable version)