robots

A human skin for robots

Another step toward human-like robots

A new material developed by scientists of the University of Tokyo appears like a skin texture, is water-resistant, and can ‘heal’ itself with a collagen plaster. The idea is to use it as a skin for humanoid robots.

As reported, the process for making it was published in the journal Matter, and it entails dipping a robot finger into a solution of collagen and human dermal fibroblasts, the two primary components that make up the connective tissue that makes it look like human skin.

The goal of developing biohybrid robotics is to make them look like ‘humans’. This is because many robots are designed to engage with people in the healthcare and service industries, who are more comfortable interacting with them if they appear to be human-like. According to the researchers, this can increase human-robot communication and even evoke likeability.

Artificial skins now on the market are constructed of silicone, which can resemble human skin in appearance, but they are not the same thing when it comes to finer textures like wrinkles. Silicone can’t also perform skin-specific activities like sweating or self-healing, and it’s difficult to conform to dynamic things with uneven surfaces.

The skin’s capacity to flex and stretch worked after healing from a wound.
Staining is visible on frozen sections of synthetic skin tissue. The cells’ nuclei were colored violet, while the extracellular matrix and cytoplasm were colored pink. This shows how an epidermal layer created a homogeneous layer that seamlessly covered the dermis equivalent’s surface.
The skin was tested by punching a ‘wound’ into it, then coating it with a collagen film and placing it on a robotic finger.
Images of skin fibroblast cells migrating onto a grafted collagen sheet used to treat a wound on artificial skin. (i) Three days after sheet application; (ii) Seven days following sheet application. It was similar to what was seen at wound sites on human skin.
Figures depict the skin-equivalent fabrication method utilized to cover the robotic finger.

To efficiently cover surfaces with skin cells, they created a tissue molding approach to physically shape skin tissue around the robot, which resulted in seamless skin coverage. To create the skin, the team submerged a robotic finger in a solution of collagen and human dermal fibroblasts.

Collagen helps to maintain the skin’s flexibility and strength, whereas fibroblast cells are important for hair growth and wound healing. This combination allows the artificial skin to shrink organically around the hardware, creating a homogeneous foundation for the following layer.

The team next applied human epidermal keratinocytes to the skin, which form about 90% of the outermost layer of human skin and have skin-like texture and moisture-retaining barrier properties.

The skin had enough strength and elasticity to keep the robotic finger intact when it was coiled, bent, and stretched, according to scientists and engineers. The top layer could be lifted with tweezers, and any wounds would heal on their own if wrapped with a collagen bandage. The bandage melded into the skin over time and was able to withstand repetitive joint motions.

The finger skin also repelled water. However, more work is needed to improve the artificial skin’s robustness and enable it to endure for long periods without feeding or waste disposal.

But that’s not all. Sensory neurons, hair follicles, nails, and sweat glands are among the more sophisticated functional systems under the skin that the team hopes to include in the future.

This is another step forward to more human-like robots. Although the intent is to engage people, especially in health facilities, and make people more at ease, this doesn’t preclude using them in other fields. As we have already explained, human-like robots are not a good thing. Besides the uncanny valley, there’s the risk that they may be used to deceive people. When they look perfectly like humans and appear indistinguishable from them, it will be easier to use them to manipulate and/or harm people.

Dan Brokenhouse

Recent Posts

Does TikTok read your mind?

An in-depth look at how TikTok's algorithm shapes user experiences, and the importance of human…

1 day ago

AI ‘ghost’ avatars

Experts warn AI "ghost" avatars could disrupt the grieving process, leading to stress, confusion, and…

1 week ago

How LLMs retrieve some stored knowledge

Large language models use linear functions to retrieve factual knowledge, providing insights into their inner…

2 weeks ago

OpenAI-powered humanoid robot

Exploring the fascinating yet concerning integration of ChatGPT into a humanoid robot by Figure AI,…

3 weeks ago

LLMs can predict the future

A new study shows large language models trained on human predictions can forecast future events…

4 weeks ago

AI superintelligence may be an ‘existential catastrophe’

A new study challenges the idea that advanced AI systems can be controlled, warning of…

1 month ago