ABSTRACT
Untethered small-scale soft robots have promising applications in minimally invasive surgery, targeted drug delivery, and bioengineering applications as they can directly and non-invasively access confined and hard-to-reach spaces in the human body. For such potential biomedical applications, the adaptivity of the robot control is essential to ensure the continuity of the operations, as task environment conditions show dynamic variations that can alter the robot's motion and task performance. The applicability of the conventional modeling and control methods is further limited for soft robots at the small-scale owing to their kinematics with virtually infinite degrees of freedom, inherent stochastic variability during fabrication, and changing dynamics during real-world interactions. To address the controller adaptation challenge to dynamically changing task environments, we propose using a probabilistic learning approach for a millimeter-scale magnetic walking soft robot using Bayesian optimization (BO) and Gaussian processes (GPs). Our approach provides a data-efficient learning scheme by finding the gait controller parameters while optimizing the stride length of the walking soft millirobot using a small number of physical experiments. To demonstrate the controller adaptation, we test the walking gait of the robot in task environments with different surface adhesion and roughness, and medium viscosity, which aims to represent the possible conditions for future robotic tasks inside the human body. We further utilize the transfer of the learned GP parameters among different task spaces and robots and compare their efficacy on the improvement of data-efficient controller learning.
ABSTRACT
Magnetically actuated miniature soft robots are capable of programmable deformations for multimodal locomotion and manipulation functions, potentially enabling direct access to currently unreachable or difficult-to-access regions inside the human body for minimally invasive medical operations. However, magnetic miniature soft robots are so far mostly based on elastomers, where their limited deformability prevents them from navigating inside clustered and very constrained environments, such as squeezing through narrow crevices much smaller than the robot size. Moreover, their functionalities are currently restricted by their predesigned shapes, which is challenging to be reconfigured in situ in enclosed spaces. Here, we report a method to actuate and control ferrofluid droplets as shape-programmable magnetic miniature soft robots, which can navigate in two dimensions through narrow channels much smaller than their sizes thanks to their liquid properties. By controlling the external magnetic fields spatiotemporally, these droplet robots can also be reconfigured to exhibit multiple functionalities, including on-demand splitting and merging for delivering liquid cargos and morphing into different shapes for efficient and versatile manipulation of delicate objects. In addition, a single-droplet robot can be controlled to split into multiple subdroplets and complete cooperative tasks, such as working as a programmable fluidic-mixing device for addressable and sequential mixing of different liquids. Due to their extreme deformability, in situ reconfigurability and cooperative behavior, the proposed ferrofluid droplet robots could open up a wide range of unprecedented functionalities for lab/organ-on-a-chip, fluidics, bioengineering, and medical device applications.
ABSTRACT
Shape-morphing magnetic soft machines are highly desirable for diverse applications in minimally invasive medicine, wearable devices, and soft robotics. Despite recent progress, current magnetic programming approaches are inherently coupled to sequential fabrication processes, preventing reprogrammability and high-throughput programming. Here, we report a high-throughput magnetic programming strategy based on heating magnetic soft materials above the Curie temperature of the embedded ferromagnetic particles and reorienting their magnetic domains by applying magnetic fields during cooling. We demonstrate discrete, three-dimensional, and reprogrammable magnetization with high spatial resolution (~38 µm). Using the reprogrammable magnetization capability, reconfigurable mechanical behavior of an auxetic metamaterial structure, tunable locomotion of a surface-walking soft robot, and adaptive grasping of a soft gripper are shown. Our approach further enables high-throughput magnetic programming (up to 10 samples/min) via contact transfer. Heat-assisted magnetic programming strategy described here establishes a rich design space and mass-manufacturing capability for development of multiscale and reprogrammable soft machines.