Braille is a system of raised dots that allows blind and visually impaired people to read and write. It is also a challenging task for robots, as it requires high sensitivity and accuracy to detect the tiny changes in the surface of the paper. However, researchers from the University of Cambridge have developed a robotic sensor that can read braille at speeds roughly double that of most human readers, using artificial intelligence techniques.
The research team, led by Professor Fumiya Iida, used machine learning algorithms to teach a robotic sensor to quickly slide over lines of braille text, instead of touching each letter pattern individually, as most existing robotic braille readers do. The robot was able to read the braille at 315 words per minute at close to 90% accuracy, surpassing the average human reading speed of 150 words per minute.
The robotic sensor has a camera in its ‘fingertip’, which captures images of the braille dots as it moves over them. The images are then processed by a neural network, which recognizes the patterns and converts them into text. The robot also adjusts its speed and force according to the feedback from the sensors, to optimize its reading performance.
The robot braille reader was not developed as an assistive technology, but as a way to test the capabilities of robotic hands or prosthetics with comparable sensitivity to human fingertips. The researchers say that the high sensitivity required to read braille makes it an ideal benchmark for the development of such devices, which could have applications in various fields, such as medicine, manufacturing, or agriculture.
The following table summarizes some of the key features and advantages of the robot braille reader compared to human readers and existing robotic readers:
| Feature | Robot braille reader | Human reader | Existing robotic reader |
|---|---|---|---|
| Reading speed | 315 words per minute | 150 words per minute | 20 words per minute |
| Reading accuracy | 87% | 95% | 99% |
| Reading method | Sliding over lines | Sliding over lines | Touching each letter |
| Energy efficiency | High | High | Low |
| Adaptability | High | High | Low |
The researchers hope that their robot braille reader will inspire more research and innovation in the field of robotics, especially in developing soft and sensitive robot hands that can mimic human skills. They also plan to improve the robot’s accuracy and robustness, and to explore other applications of the sensor, such as reading other tactile codes or textures.
