Trend AnalysisOther Engineering
Self-Healing Robots: Artificial Muscles That Repair Themselves
Biological muscles can heal from damage, adjust their stiffness, and respond to multiple stimuli. Achieving all three in artificial systems has been a longstanding challenge. Recent advances in self-healing polymers and bio-inspired designs are bringing artificial muscles closer to biological performance.
By Sean K.S. Shin
This blog summarizes research trends based on published paper abstracts. Specific numbers or findings may contain inaccuracies. For scholarly rigor, always consult the original papers cited in each post.
Biological muscle is a remarkable material: it contracts and relaxes with precise control, adjusts its stiffness depending on the task, responds to chemical and electrical stimuli, and can heal from moderate damage. Replicating even a subset of these capabilities in artificial systems would transform soft roboticsโenabling robots that are not merely flexible but adaptive, resilient, and self-maintaining.
The Research Landscape
Muscle-Like Actuation with Self-Healing
Huang, Wu, and Li (2024), with 41 citations in Advanced Materials, present the most significant materials advance: self-healing actuator materials that achieve muscle-like contraction and relaxation through multilevel molecular relaxations. The material combines two properties that have historically been difficult to unite: mechanical strength (necessary for effective actuation) and self-healing (which typically requires soft, mobile molecular networks).
The solution is a hierarchically structured polymer where different molecular segments handle different functions: rigid segments provide mechanical strength, while dynamic bonds (hydrogen bonds and metal-ligand coordination) provide self-healing capability. When the material is damaged (cut or punctured), the dynamic bonds at the damage site reform over time, restoring 85-95% of original mechanical properties within 24 hours at room temperature.
Intelligent Damage Detection
Krings, McManigal, and Markvicka (2025) address a complementary challenge: detecting that damage has occurred in the first place. A self-healing material that cannot detect damage cannot initiate repair. Their system embeds resistive sensing elements within soft robotic actuators, enabling real-time detection of puncture damage through changes in electrical resistance.
The system detects punctures as small as 1mm diameter within seconds, and the location accuracy is sufficient to guide repair mechanisms. The challenge is robustness: sensing elements must remain functional during the extreme deformations that soft robots undergo during normal operationโa requirement that rigid sensors cannot meet.
Multi-Stimulus Response
Must and Vihmar (2025) explore actuators that respond to multiple stimuliโnot just electrical signals (as conventional motors do) but also temperature, light, humidity, and chemical signals. Bio-inspired designs that leverage high-performance polymers and ionic electroactive materials can produce actuators that bend toward a light source, stiffen in response to temperature, and contract in response to chemical signalsโapproaching the multi-modal responsiveness of biological tissue.
Variable Stiffness
Li, Bai, and Wang (2025) draw inspiration from the octopus armโa boneless limb that can be simultaneously strong and flexible. Their hydrostatic artificial muscle (HAM) achieves variable stiffness by adjusting the internal pressure of an incompressible fluid within a flexible shell. At low pressure, the actuator is soft and compliant; at high pressure, it becomes rigid enough to grip and manipulate objects.
Critical Analysis: Claims and Evidence
<
| Claim | Evidence | Verdict |
|---|
| Self-healing actuators can recover 85-95% of original properties | Huang et al.'s materials testing | โ
Supported โ 41 citations |
| Resistive sensing can detect mm-scale punctures in soft actuators | Krings et al.'s sensing experiments | โ
Supported โ demonstrated in lab conditions |
| Multi-stimulus actuators approach biological responsiveness | Must et al.'s design demonstrations | โ ๏ธ Uncertain โ individual stimuli work; integrated multi-stimulus systems are early |
| Variable stiffness is achievable through hydrostatic design | Li et al.'s HAM prototype | โ
Supported โ demonstrated |
What This Means for Your Research
For materials scientists, the self-healing actuator from Huang et al. demonstrates that strength and healing are not inherently opposedโthe solution is hierarchical molecular design. For roboticists, the combination of self-healing materials with damage sensing (Krings et al.) provides the building blocks for autonomous self-maintenance in soft robots.
Explore related work through ORAA ResearchBrain.
๋ฉด์ฑ
์กฐํญ: ์ด ๊ฒ์๋ฌผ์ ์ ๋ณด ์ ๊ณต์ ๋ชฉ์ ์ผ๋ก ํ ์ฐ๊ตฌ ๋ํฅ ๊ฐ์์ด๋ค. ํ์ ์ฐ๊ตฌ์์ ์ธ์ฉํ๊ธฐ ์ ์ ํน์ ์ฐ๊ตฌ ๊ฒฐ๊ณผ, ํต๊ณ, ์ฃผ์ฅ์ ์๋ณธ ๋
ผ๋ฌธ์ ํตํด ๊ฒ์ฆํด์ผ ํ๋ค.
์๊ฐ ์น์ ๋ก๋ด: ์ค์ค๋ก ๋ณต๊ตฌํ๋ ์ธ๊ณต ๊ทผ์ก
์๋ฌผํ์ ๊ทผ์ก์ ๋๋ผ์ด ์์ฌ์ด๋ค. ์ ๋ฐํ๊ฒ ์ ์ด๋๋ฉฐ ์์ถ๊ณผ ์ด์์ ๋ฐ๋ณตํ๊ณ , ์์
์ ๋ฐ๋ผ ๊ฐ์ฑ์ ์กฐ์ ํ๋ฉฐ, ํํ์ ยท์ ๊ธฐ์ ์๊ทน์ ๋ฐ์ํ๊ณ , ์ค๊ฐ ์ ๋์ ์์์ผ๋ก๋ถํฐ ์ค์ค๋ก ์น์ ํ ์ ์๋ค. ์ด๋ฌํ ํน์ฑ๋ค ์ค ์ผ๋ถ๋ง์ด๋ผ๋ ์ธ๊ณต ์์คํ
์์ ๊ตฌํํ ์ ์๋ค๋ฉด ์ํํธ ๋ก๋ด๊ณตํ์ ํฌ๊ฒ ๋ณํํ ๊ฒ์ด๋ค. ๋จ์ํ ์ ์ฐํ ๋ก๋ด์ ๋์ด, ์ ์์ ์ด๊ณ ํ๋ ฅ์ ์ด๋ฉฐ ์๊ฐ ์ ์ง๊ฐ ๊ฐ๋ฅํ ๋ก๋ด์ ๋ง๋ค ์ ์๊ฒ ๋๋ค.
์ฐ๊ตฌ ๋ํฅ
๊ทผ์กํ ๊ตฌ๋๊ณผ ์๊ฐ ์น์
Huang, Wu, Li (2024)๋ Advanced Materials์ ๊ฒ์ฌ๋ ๋
ผ๋ฌธ์์ 41ํ ํผ์ธ์ฉ์ ๊ธฐ๋กํ๋ฉฐ, ์์ฌ ๋ถ์ผ์์ ๊ฐ์ฅ ์ฃผ๋ชฉํ ๋งํ ์ง์ ์ ์ ์ํ๋ค. ์ด๋ค์ด ๊ฐ๋ฐํ ์๊ฐ ์น์ ๊ตฌ๋ ์์ฌ๋ ๋ค์ธต์ ๋ถ์ ์ด์์ ํตํด ๊ทผ์ก๊ณผ ์ ์ฌํ ์์ถ ๋ฐ ์ด์์ ๊ตฌํํ๋ค. ์ด ์์ฌ๋ ์ญ์ฌ์ ์ผ๋ก ๊ฒฐํฉํ๊ธฐ ์ด๋ ค์ ๋ ๋ ๊ฐ์ง ํน์ฑ, ์ฆ ํจ๊ณผ์ ์ธ ๊ตฌ๋์ ํ์ํ ๊ธฐ๊ณ์ ๊ฐ๋์ ์ผ๋ฐ์ ์ผ๋ก ์ ์ฐํ๊ณ ์ด๋์ฑ ์๋ ๋ถ์ ๋คํธ์ํฌ๋ฅผ ์๊ตฌํ๋ ์๊ฐ ์น์ ๋ฅ๋ ฅ์ ํจ๊ป ๊ฐ์ถ๊ณ ์๋ค.
ํด๊ฒฐ์ฑ
์ ๊ณ์ธต์ ์ผ๋ก ๊ตฌ์กฐํ๋ ํด๋ฆฌ๋จธ๋ก, ์๋ก ๋ค๋ฅธ ๋ถ์ ์ธ๊ทธ๋จผํธ๊ฐ ๊ฐ๊ธฐ ๋ค๋ฅธ ๊ธฐ๋ฅ์ ๋ด๋นํ๋ค. ๊ฐ์ฑ ์ธ๊ทธ๋จผํธ๋ ๊ธฐ๊ณ์ ๊ฐ๋๋ฅผ ์ ๊ณตํ๊ณ , ๋์ ๊ฒฐํฉ(์์ ๊ฒฐํฉ ๋ฐ ๊ธ์-๋ฆฌ๊ฐ๋ ๋ฐฐ์)์ ์๊ฐ ์น์ ๋ฅ๋ ฅ์ ๋ถ์ฌํ๋ค. ์์ฌ๊ฐ ์์(์ ๋จ ๋๋ ์ฒ๊ณต)๋๋ฉด, ์์ ๋ถ์์ ๋์ ๊ฒฐํฉ์ด ์๊ฐ์ด ์ง๋๋ฉด์ ์ฌํ์ฑ๋์ด ์ค์จ์์ 24์๊ฐ ์ด๋ด์ ์๋ ๊ธฐ๊ณ์ ํน์ฑ์ 85-95%๋ฅผ ํ๋ณตํ๋ค.
์ง๋ฅํ ์์ ๊ฐ์ง
Krings, McManigal, Markvicka (2025)๋ ์๋ณด์ ์ธ ๊ณผ์ , ์ฆ ์์์ด ๋ฐ์ํ์์ ์ต์ด๋ก ๊ฐ์งํ๋ ๋ฌธ์ ๋ฅผ ๋ค๋ฃฌ๋ค. ์์์ ๊ฐ์งํ์ง ๋ชปํ๋ ์๊ฐ ์น์ ์์ฌ๋ ๋ณต๊ตฌ๋ฅผ ์์ํ ์ ์๋ค. ์ด๋ค์ ์์คํ
์ ์ํํธ ๋ก๋ด ๊ตฌ๋๊ธฐ ๋ด๋ถ์ ์ ํญ์ฑ ๊ฐ์ง ์์๋ฅผ ๋ด์ฅํ์ฌ, ์ ๊ธฐ ์ ํญ์ ๋ณํ๋ฅผ ํตํด ์ฒ๊ณต ์์์ ์ค์๊ฐ์ผ๋ก ๊ฐ์งํ๋ค.
์ด ์์คํ
์ ์ง๊ฒฝ 1mm ์ดํ์ ์ฒ๊ณต์ ์ ์ด ์ด๋ด์ ๊ฐ์งํ๋ฉฐ, ์์น ์ ํ๋๋ ๋ณต๊ตฌ ๋ฉ์ปค๋์ฆ์ ์๋ดํ๊ธฐ์ ์ถฉ๋ถํ ์์ค์ด๋ค. ํต์ฌ ๊ณผ์ ๋ ๊ฒฌ๊ณ ์ฑ์ด๋ค. ๊ฐ์ง ์์๋ ์ํํธ ๋ก๋ด์ด ์ ์ ์๋ ์ค์ ๊ฒช๋ ๊ทน๋จ์ ์ธ ๋ณํ ๋์์๋ ๊ธฐ๋ฅ์ ์ ์งํด์ผ ํ๋๋ฐ, ์ด๋ ๊ฐ์ฑ ์ผ์๋ก๋ ์ถฉ์กฑํ๊ธฐ ์ด๋ ค์ด ์๊ฑด์ด๋ค.
๋ค์ค ์๊ทน ๋ฐ์
Must์ Vihmar (2025)๋ ๊ธฐ์กด ๋ชจํฐ์ฒ๋ผ ์ ๊ธฐ ์ ํธ์๋ง ๋ฐ์ํ๋ ๊ฒ์ด ์๋๋ผ ์จ๋, ๋น, ์ต๋, ํํ ์ ํธ ๋ฑ ๋ค์ํ ์๊ทน์ ๋ฐ์ํ๋ ๊ตฌ๋๊ธฐ๋ฅผ ์ฐ๊ตฌํ๋ค. ๊ณ ์ฑ๋ฅ ํด๋ฆฌ๋จธ์ ์ด์จ์ฑ ์ ๊ธฐํ์ฑ ์์ฌ๋ฅผ ํ์ฉํ ์์ฒด ๋ชจ๋ฐฉ ์ค๊ณ๋ฅผ ํตํด, ๊ด์์ ํฅํด ๊ตฝํ๊ณ , ์จ๋์ ๋ฐ์ํ์ฌ ๊ฒฝ์ง๋๋ฉฐ, ํํ ์ ํธ์ ๋ฐ์ํ์ฌ ์์ถํ๋ ๊ตฌ๋๊ธฐ๋ฅผ ๊ตฌํํ ์ ์๋ค. ์ด๋ ์๋ฌผํ์ ์กฐ์ง์ ๋ค์ค ๋ชจ๋ฌ ๋ฐ์์ฑ์ ๊ทผ์ ํ๋ ์์ค์ด๋ค.
๊ฐ๋ณ ๊ฐ์ฑ
Li, Bai, Wang (2025)์ ๋ผ๊ฐ ์์์๋ ๊ฐํ๊ณ ์ ์ฐํ ์ฑ์ง์ ๋์์ ์ง๋ ๋ฌธ์ด ํ์์ ์๊ฐ์ ์ป์๋ค. ์ด๋ค์ด ๊ฐ๋ฐํ ์ ์์ ์ธ๊ณต ๊ทผ์ก(HAM, Hydrostatic Artificial Muscle)์ ์ ์ฐํ ์ธํผ ๋ด๋ถ์ ๋น์์ถ์ฑ ์ ์ฒด์ ๋ด๋ถ ์๋ ฅ์ ์กฐ์ ํ์ฌ ๊ฐ๋ณ ๊ฐ์ฑ์ ๊ตฌํํ๋ค. ์ ์ ์ํ์์๋ ๊ตฌ๋๊ธฐ๊ฐ ์ ์ฐํ๊ณ ์์์ ์ด๋ฉฐ, ๊ณ ์ ์ํ์์๋ ๋ฌผ์ฒด๋ฅผ ํ์งํ๊ณ ์กฐ์ํ ์ ์์ ๋งํผ ์ถฉ๋ถํ ๊ฒฝ์ง๋๋ค.
๋นํ์ ๋ถ์: ์ฃผ์ฅ๊ณผ ๊ทผ๊ฑฐ
<
| ์ฃผ์ฅ | ๊ทผ๊ฑฐ | ํ์ |
|---|
| ์๊ฐ ์น์ ๊ตฌ๋๊ธฐ๊ฐ ์๋ ํน์ฑ์ 85-95%๋ฅผ ํ๋ณตํ ์ ์๋ค | Huang ๋ฑ์ ์์ฌ ์ํ | โ
์ง์ง๋จ โ 41ํ ํผ์ธ์ฉ |
| ์ ํญ์ฑ ๊ฐ์ง๋ฅผ ํตํด ์ํํธ ๊ตฌ๋๊ธฐ์ mm ๋จ์ ์ฒ๊ณต์ ๊ฐ์งํ ์ ์๋ค | Krings ๋ฑ์ ๊ฐ์ง ์คํ | โ
์ง์ง๋จ โ ์คํ์ค ์กฐ๊ฑด์์ ์
์ฆ |
| ๋ค์ค ์๊ทน ๊ตฌ๋๊ธฐ๊ฐ ์๋ฌผํ์ ๋ฐ์์ฑ์ ๊ทผ์ ํ๋ค | Must ๋ฑ์ ์ค๊ณ ์์ฐ | โ ๏ธ ๋ถํ์ค โ ๊ฐ๋ณ ์๊ทน์ ์๋ํ๋, ํตํฉ๋ ๋ค์ค ์๊ทน ์์คํ
์ ์ด๊ธฐ ๋จ๊ณ |
| Variable stiffness is achievable through hydrostatic design | Li et al.'s HAM prototype | โ
Supported โ demonstrated |
What This Means for Your Research
For materials scientists, the self-healing actuator from Huang et al. demonstrates that strength and healing are not inherently opposedโthe solution is hierarchical molecular design. For roboticists, the combination of self-healing materials with damage sensing (Krings et al.) provides the building blocks for autonomous self-maintenance in soft robots.
Explore related work through ORAA ResearchBrain.
References (5)
[1] Huang, Z., Wu, Z., & Li, C. (2024). Self-Healing Yet Strong Actuator Materials with Muscle-Like Diastole and Contraction. Advanced Materials.
[2] Krings, E.J., McManigal, P., & Markvicka, E.J. (2025). Intelligent Self-Healing Artificial Muscle: Mechanisms for Damage Detection and Autonomous Repair. Proc. IEEE ICRA 2025.
[3] Must, I., Valdur, K.-A., & Vihmar, M. (2025). Multi-stimulus-responsive soft actuators: bio-inspired designs and intelligent material systems. Proc. SPIE.
[4] Li, K., Bai, C., & Wang, J. (2025). Octopus-Inspired Hydrostatic Artificial Muscle with Variable Stiffness. Proc. RoboSoft 2025, IEEE.
Must, I., Valdur, K. A., Vihmar, M., Sarokin, Y., Chang, L., & Aabloo, A. (2025). Multi-stimulus-responsive soft actuators: integrating bio-inspired designs and intelligent material systems. Electroactive Polymer Actuators, Sensors, and Devices (EAPAD) 2025, 45.