Insider Brief
- Researchers at Japan Advanced Institute of Science and Technology, backed by Japan’s KAKENHI and JST PRESTO grants, developed ProTac, a vision-based soft robotic skin combining touch and proximity sensing.
- The system uses a polymer-dispersed liquid crystal layer with embedded cameras that switch between transparent and opaque states, enabling detection of nearby objects and registration of touch and pressure in real time.
- Validated through a prototype robot arm, ProTac demonstrated adaptive safety behaviors and has potential applications in factories, elder care, agriculture, and humanoid robots requiring safe, responsive interaction.
Japanese researchers have developed a soft robotic skin that combines touch and proximity sensing, supported by national Japan Society for the Promotion of Science research grants.
A team at the Japan Advanced Institute of Science and Technology, led by Professor Van Anh Ho and Dr. Quan Khanh Luu, created ProTac, a vision-based sensing layer published in IEEE Transactions on Robotics. The work received support from Japan’s Grant-in-Aid for Scientific Research (KAKENHI) and the JST PRESTO program.
According to JAIST, ProTac uses a polymer-dispersed liquid crystal (PDLC) layer that can switch between transparent and opaque states under voltage. When transparent, cameras embedded behind the skin detect nearby objects. When opaque, the cameras measure how the material deforms to register touch, pressure, and point of contact. This dual-mode design reduces complexity while offering large-area, real-time perception.
“This innovation with sensor design simplicity allows robots to perceive both contact and nearby obstacles across a large-area skin in real time, which is difficult to achieve with conventional electronic skins,” noted Ho.
The researchers said they validated their design by creating a prototype called the ProTac link, a cylindrical robot arm segment covered in the sensing skin and outfitted with stereo cameras. The device could detect multiple approaching objects, calculate distance, and track several contact points simultaneously. It also demonstrated adaptive safety behaviors such as slowing in response to proximity and avoiding unexpected collisions.
According to the study, ProTac could enhance collaborative robots in factories, assistive service robots in elder care, and soft robotic systems in agriculture or domestic environments. In manufacturing, the skin would allow robots to automatically reduce speed when workers approach. In health care, robots could maintain delicate, responsive contact when helping elderly users with mobility.
“ProTac can be applied to dexterous robotic manipulation in various domains where safety and delicate physical interaction are critical,” Luu pointed out.
The system also includes machine learning algorithms and control strategies to convert raw sensor data into robot actions. Designed for modularity, the technology can be integrated into both new robotic systems and retrofitted onto existing platforms.
While ProTac is still at the prototype stage, the researchers emphasize its scalability and cost-effectiveness compared to complex electronic skins. They also released design files and software as open source to accelerate adoption by the robotics community. The team sees this work as an early step toward full-body multimodal sensing for humanoid robots, a capability they argue will be essential for general-purpose machines that can operate safely in human environments.
“In the long term, this research lays the groundwork for intelligent general-purpose humanoid robots equipped with full-body multimodal soft sensing skins, especially those with a sense of touch,” said Ho..
