We observe that the self-organization of many simple and deformable elements in a bottom-up manner creates a rich variety of physical adaptation in nature. These examples exist on multiple scale lengths from embryonic cells to fish schools. Conversely, we build robotic mechanisms following pre-defined designs and optimize them for certain tasks with conventional top-down design tools.
I believe that we can build novel physically adaptive robots by combining these two complementary design approaches. Inspirations from biology, soft deformable materials, and self-assembling systems may allow us to explore unprecedented forms and functions in a bottom-up manner. We can guide this exploration with top-down control and learning methods, and achieve repeatability and programmability in short time-scales. I believe that once a robot is physically adaptive, increments to its computational intelligence will always have multiplying effects on its adaptivity, durability, and autonomy.
Bio-Inspired Soft Robots
The depiction of a robot used to be not much different than an aluminum toaster for a long time. Inspirations from biology changed this outlook and introduced soft and deformable materials to robot designs. With the advances in materials research, now we can easily fabricate and functionalize such materials and build robots that demonstrate tasks that were not possible with the conventional industrial robots.
I always look at biology to find inspirations for the morphological design and functional variety for my robots. I benefit from the viscoelastic deformation of soft materials to build robots that demonstrate examples of bio-inspired sensing, locomotion, and manipulation. Some of my projects include a mobile robot that produces spider-inspired draglines, a robot arm that fabricates soft sensors to detect different physical features, and anthropomorphic robotic fingers that can handle a wide range of objects from chopsticks to lettuce.
I am currently working on the sub-centimeter miniature scale and designing magnetically actuated soft robots that can assemble in different configurations to perform multi-modal locomotion and manipulation.
You can find some of the projects I worked on here:
Sensing: Sensors (2014), IROS (2014), PLoS One (2013)
Manipulation: Soft Robotics (2017), Frontiers (2016), Bio&Bio (2016), AMAM (2016)
Locomotion:Bio&Bio (2014), ICRA (2014), IROS (2013), AMAM (2013), ICRA (2011)
Self-assembly is a ubiquitous process in both of the living and non-living systems that generate complex and functional structures from the local interactions between a large set of simpler components. Especially in biology, physical adaptation is explained by linking the emergence of functions (e.g., cell motility, cell division, and morphogenesis) to the formation of such structures as a result of the collective behavior of neighboring cells.
The ability to program (i.e., to control) the self-assembly process is very exciting for the robotics field as it may enable alternative design methods to build novel robotic systems. This is also an interdisciplinary line of research that will additionally elucidate fundamental physics and enable the fabrication of novel functional materials.
I am currently working on the design of physical mechanisms that allow the programming of the self-assembly pathway of miniature soft robots. These mechanisms enable the self-assembly of different functional robotic structures in a repeatable fashion.
Here are the projects I am currently working on:
Reprogrammable self-assembly: PNAS (2020).
Soft robotic self-assembly: Coming soon.
Soft robots have great potential in achieving bio-inspired functions involving viscoelastic deformations. However, the same features that enable this potential such as having virtually infinite degrees of freedom, soft continuum body structures, and material-enabled behavior, make it challenging to model and predict their kinematics and dynamics, and apply conventional robot control methods. Also, common approximation methods depending on body discretization and constant curvature assumptions fall short of revealing the true potential of these robots.
I believe that model-free machine-learning methods may help us explore the vast design and function space of soft robots without enforcing pre-conceived kinematic or dynamic models. However, such an approach requires physical experiments that guarantee repeatability, high precision, and reliable feedback information. Once we have these experimental platforms, we can explore unprecedented robotic functions and designs, and find the missing the correlation between the control and state space of soft robots.
My current work focuses on designing physical experiments with millimeter-scale magnetically actuated soft robots and find the controller parameters that generate bio-inspired walking or manipulation.
Here are the projects I am currently working on: