Beijing, March 26 (NationPress) Envision a robotic hand that synchronizes its palm and fingers with remarkable human-like accuracy -- not only grasping items with impeccable stability but also detecting their 3D contours and identifying different materials.

Scientists from Shanghai Jiao Tong University have turned this vision into reality, addressing a critical challenge in the field of robotics, as reported by the Xinhua news agency.

While soft robotic hands equipped with sensing abilities have shown potential, current designs have primarily concentrated on finger sensitivity, overlooking the essential role of the palm in manipulation.

The palm, a significant contact area that provides mechanical support and sensory feedback, was previously underexplored due to limited sensing density and interaction with the fingers.

The research team successfully tackled these obstacles by creating a fully integrated system where the palm and fingers function in perfect synchronization, achieving unparalleled levels of sensitivity and control.

Central to this breakthrough are three innovative features: a high-resolution tactile palm boasting 181,000 sensing units per square centimeter (754 times more sensitive than human skin), highly flexible fingers capable of bending 204.3 degrees, and smart coordination algorithms that facilitate smooth palm-finger interaction.

The palm's advanced visual-tactile system merges micro-cameras with multi-layer elastic sensors, while the fiber-reinforced pneumatic fingers deliver both strength and finesse with a grasping force of 14.6 newtons.

This integration empowers the robotic hand to execute tasks with exceptional human-like ability. In demonstrations, it performed intricate maneuvers such as retrieving cards with pinpoint precision, identifying microscopic defects in fabrics, and even pouring tea while automatically adjusting its grip to avert spills.

The research team has augmented the system's capabilities by incorporating cutting-edge AI technology, significantly enhancing its object recognition proficiency. In challenging blind tests, the robotic hand showcased remarkable accuracy by identifying and retrieving specific target items from opaque bags filled with assorted small objects -- ranging from tiny peas and sunflower seeds to larger nuts and bolts.

Remarkably, the system achieved an overall recognition accuracy of 88 percent, reaching 100 percent accuracy for distinctive objects like sunflower seeds.

"Our findings indicate the potential of merging rich palm tactile sensing with soft dexterous fingers for advanced interactive robotic functions," the researchers remarked.

The study is published in the journal Nature Communications.