Embodying Physical Computing into Soft Robots

TL;DR

This work articulates a framework for embedding physical computing in soft robots to achieve robust, intelligent behavior without reliance on CMOS electronics. It categorizes embodied computation into analog oscillator-based rhythmic control, analog physical reservoir computing for perception and control, and algorithmic mechanical computing using bistable logic and fluidic circuits, with illustrative demonstrations of locomotion, payload classification, and memory. The paper surveys state-of-the-art examples and articulates a roadmap for future advances, including higher-density kernels, distributed computation across soft bodies, and hybrid mechanical-electrical circuits. Together, these approaches offer a path toward fully mechanical intelligence in soft robotics, combining material science, metamaterials, and computing theory to enable memory, decision-making, and adaptive behavior directly in the robot body.

Abstract

Softening and onboarding computers and controllers is one of the final frontiers in soft robotics towards their robustness and intelligence for everyday use. In this regard, embodying soft and physical computing presents exciting potential. Physical computing seeks to encode inputs into a mechanical computing kernel and leverage the internal interactions among this kernel's constituent elements to compute the output. Moreover, such input-to-output evolution can be re-programmable. This perspective paper proposes a framework for embodying physical computing into soft robots and discusses three unique strategies in the literature: analog oscillators, physical reservoir computing, and physical algorithmic computing. These embodied computers enable the soft robot to perform complex behaviors that would otherwise require CMOS-based electronics -- including coordinated locomotion with obstacle avoidance, payload weight and orientation classification, and programmable operation based on logical rules. This paper will detail the working principles of these embodied physical computing methods, survey the current state-of-the-art, and present a perspective for future development.

Figures (5)

• Figure 1: The architecture and categorization of physical computing. (a) The computing architecture adopted in this perspective includes input encoding, output decoding, and programmable input-output evolution. (b-d) The two types of physical computing are analog (up) and algorithmic (bottom). (b) The early attempts at analog and algorithmic computing were purely mechanical (the Harmonic Analyzer photo adapted from hammack2014albert with permission, and the Difference Machine photo credits to Science Museum London, CC Attribution-SA 2.0). (c) The electronic implementations of these two computing approaches have become pillars of our modern life. (d) Examples of physical computing embodied in soft robots, including, from top to bottom, an electronics-free legged robot with an analog oscillator drotman2021electronics (image credit to David Baillot, Jacobs School of Engineering, UC San Diego), a modular manipulator with embodied reservoir computing (image adapted from wang2025re CC BY 4.0), and a robotic hand with fluidic logic control (image adaptive from jiao2024reprogrammable with permission).
• Figure 2: Analog oscillators and rhythmic deformation. (a) Electronics-free pneumatic control: A soft ring-oscillator circuit generates rhythmic leg actuation from a constant pressure input, enabling a quadruped to produce diagonal-couplet walking gaits. A bistable 4/2 switch selects gait direction, and dual oscillators set the phase between leg pairs (image adapted from drotman2021electronics with permission). (b) Controller-free SMA modular robot: A curved mono stable beam and a mechanical slider can convert a single DC power supply into sustained self-oscillation. And a bistable switch can alternate power supply between the front and back modules for out-of-phase deformation and crawling (image adapted from zhou2024self with permission). (c) Twisted LCE ribbon robot: The ambient heating drives continuous self-rolling of this robotic structure for locomotion. When the robot contacts an obstacle, it will store elastic energy and then snaps to reverse its direction, enabling autonomous avoidance and maze escape (image adapted from zhao2022twisting with permission).
• Figure 3: Open and closed-loop physical reservoir computing in soft robots. (a) Open-loop PRC for information perception: A modular manipulator with embedded strain gauges is driven by SMA actuators. Its high-dimensional body dynamics (measured by the strain $s_i(t)$) serve as reservoir states, which can be processed with trained linear readout $w_i$ to decode and identify the payload (image adapted from wang2025embodied with permission). (b) Other examples of information perception with open-loop PRC, including terrain classification (image adapted from nakamura2025flexible with permission), wind detection on a compliant membrane wing (image adapted from tanaka2021flapping with permission), and a self-sensing shape memory alloy actuator that could predict its end effector trajectory (image adapted from shougat2023self with permission). (c) Closed-loop PRC for embodied control: A quadruped robot uses its compliant spine as the reservoir. The four outputs of the reservoir kernel are fed back to the leg actuators to generate robust and adaptable locomotion gaits (image adapted from 2013SpineReservoir with permission). (d) Other examples of control embodiments with closed-loop PRC, including manipulation with a multi-segment continuum arm (image adapted from eder2018morphological with permission) and a surface-swimming robot (adapted from horii2021physical with permission).
• Figure 4: Physical Algorithmic Computing. (a) Reprogrammable metamaterial processors with robotic fingers: Fluidic unit-cells with 0-1 binary states are connected to create mechanical logic circuitry to control finger actions (image adapted from jiao2024reprogrammable with permission). (b) Complementary soft pneumatic valves: Piston-based, four-terminal modules are paired to achieve Boolean logic operation, non-volatile latches, and analog pressure regulation. Then they are integrated with sub-circuits to create ring oscillators and counters to control crawling robots and wearable devices (image adapted from decker2022programmable with permission). (c) Soft-matter computer: Conductive-fluid receptors transduce spatiotemporal fluid patterns into electrical drives, which can realize analog filtering, amplification, and logic gates with simple composition. As a result, such conductive fluidic mechanism enables on-body control for locomotion, reflexive grasping, and behavior switching (image adapted from garrad2019soft with permission).
• Figure 5: Future directions for embodying physical computing in soft robots: One can continue to advance the computing kernel's density and capacity in soft robots by adopting new strategies to process encoded inputs, either locally or through a centralized kernel. To illustrate some of the promising concepts, we envision an octopus-inspired soft robot with computational capabilities distributed across its tentacles as well as in its brain, following the mechanical computing framework described in this manuscript. Each component can encode, process, and decode data (with mechanical memory reserved for storage). Starting from the top right tentacle and moving clockwise: Bistable soft shells enable rule-changeable logic operations (image adapted from yang2025bistable with permission); Information processing during transmission via non-dispersive mechanical solitary waves (image adapted from byun2024integrated with permission); A mechanical neural network offloads computation and can be attached to the robot's skin (image adapted from hopkins2023using with permission); Mechanical analog-to-digital converters can be embedded in the tentacles (image adapted from el2022mechanical with permission). In the robot's brain, miniaturized physical circuits mimic an algorithmic logic unit (ALU) (image adapted from song2019additively with permission). Finally, reprogrammable and non-volatile mechanical memories can store data either with magnetic (left, image adapted from chen2021reprogrammable with permission) or thermal principles (right, image adapted from meng2023encoding with permission).