Robotic vision taught to distinguish between types of human touch

New ShadowSense technology—detects and identifies human touch using a camera. This

universal, compared to tactile sensors, which simply take up space and add weight to the structure.

The new development arose as an idea to createtow robots that could guide people to safety during an emergency. Such a robot must be able to communicate with people in extreme conditions. 

Installation of a large number of contact sensorswould have added weight to the robot, so the developers solved the problem differently. They measured the intensity and type of tactile contact using a camera: it can analyze how a person touches the robot and what their intentions are. 

The robot prototype consists of nylon skin,stretched around a cylindrical skeleton. Under the robot's skin there is a USB camera that connects to a laptop. The researchers developed a neural network algorithm that uses previously recorded training data to distinguish between six palm touch gestures: punching, two-handed touching, hugging, pointing, and no touching.

As a result, the robot will determine the type of interaction with an accuracy of 87.5 to 96%, depending on the lighting.

Read more:

Look at an 8 trillion pixel image of Mars

A nuclear rocket engine is being built for flights to Mars. How is it dangerous?

Abortion and science: what will happen to the children who will give birth