3 December 2021 in Science Advances
A light weight, economic, and high–energy density Zn/MnO 2 fiber battery was integrated with a textile body area network.
26 April 2023 in Science Robotics
As the field of soft robotics advances, full autonomy becomes highly sought after, especially if robot motion can be powered by environmental energy. This would present a self-sustained approach in terms of both energy supply and motion control. Now, autonomous movement can be realized by leveraging out-of-equilibrium oscillatory motion of stimuli-responsive polymers under a constant light source. It would be more advantageous if environmental energy could be scavenged to power robots. However, generating oscillation becomes challenging under the limited power density of available environmental energy sources. Here, we developed fully autonomous soft robots with self-sustainability based on self-excited oscillation. Aided by modeling, we have successfully reduced the required input power density to around one-Sun level through a liquid crystal elastomer (LCE)–based bilayer structure. The autonomous motion of the low-intensity LCE/elastomer bilayer oscillator “LiLBot” under low energy supply was achieved by high photothermal conversion, low modulus, and high material responsiveness simultaneously. The LiLBot features tunable peak-to-peak amplitudes from 4 to 72 degrees and frequencies from 0.3 to 11 hertz. The oscillation approach offers a strategy for designing autonomous, untethered, and sustainable small-scale soft robots, such as a sailboat, walker, roller, and synchronized flapping wings.
28 May 2025 in Science Robotics
Achieving lifelike autonomy remains a long-term aspiration, yet soft robots so far have mostly demonstrated rudimentary physical intelligence that relies on manipulation of external stimuli to generate continuous motion. To realize autonomous physical intelligence (API) capable of self-regulated sensing, decision-making, and actuation, a promising approach is creating nonlinear time-lag feedback embedded within materials, where a constant stimulus elicits delayed responses to enable autonomous motion. This Review explores such feedback mechanisms, traces the evolution of physically intelligent robots, outlines strategies for embedding API in soft robots under diverse environments, and further discusses challenges and future directions beyond simple locomotion.
28 August 2025 in Science
Cavitation, characterized by formation of vapor bubbles in a low-pressure or high-temperature region of a liquid, is often destructive, but it can be harnessed for actuators and robots. We exploit cavitation to accumulate substantial energy in superheated liquids by suppressing its immediate release until reaching a stability limit. The energetic, unstable bubbles collapse violently, producing a burst of high power and force that initiates motion. Notably, a millimeter-scale device launched by cavitation can jump to a height of 1.5 meters—reaching a 12 meters per second (m/s) peak velocity, a 7.14 × 10 4 m/s 2 acceleration, and a 0.64% energy efficiency—and can also swim on water at 12 centimeters per second. Cavitation-based launching works with a broad range of device materials, liquid media, stimuli, and operational environments.
24 October 2024 in Nature Materials
15 August 2025 in Nature Communications
Autonomous soft robots with built-in feedback can achieve self-sustained motion under constant, untethered stimuli. However, these systems are constrained to single-mode actuation within a fixed structure under the same type of stimulus and require human intervention to switch modes, lacking the robust and efficient self-adaptation of living organisms in changing environments. Inspired by Gymnura micrura, we developed a light-responsive soft actuator with engineered asymmetry and a dynamic structure, integrating two distinct built-in feedback mechanisms governed by intrinsic bifurcation. Thus, the actuator can seamlessly switch between three different motion modes—tracking, undulation, and oscillation—exhibiting self-adaptation to environmental changes (e.g., light intensity, viscosity, temperatures, and physical contact). Furthermore, this multimodal capability facilitates unique environmental interactions, expanding applications beyond locomotion to include fluid dynamics, electronics, and environmental monitoring. Such an advancement in physical intelligence represents a pivotal step toward next-generation autonomous soft robotic systems, unlocking higher-level autonomy and unprecedented adaptive behaviors.