{"id":208574,"date":"2021-03-22T22:28:05","date_gmt":"2021-03-22T19:28:05","guid":{"rendered":"https:\/\/en.buradabiliyorum.com\/mit-algorithm-arranges-sensors-on-soft-robots-to-improve-their-operations\/"},"modified":"2021-03-22T22:28:05","modified_gmt":"2021-03-22T19:28:05","slug":"mit-algorithm-arranges-sensors-on-soft-robots-to-improve-their-operations","status":"publish","type":"post","link":"https:\/\/buradabiliyorum.com\/en\/mit-algorithm-arranges-sensors-on-soft-robots-to-improve-their-operations\/","title":{"rendered":"#MIT algorithm arranges sensors on soft robots to improve their operations"},"content":{"rendered":"

“#MIT algorithm arranges sensors on soft robots to improve their operations<\/strong>”
\n<\/p>\n

\n Soft robots can be far more flexible than\u00a0their rigid relatives, but their vast ranges of motion makes it hard to\u00a0map the location of their body parts.<\/p>\n

A new algorithm\u00a0could give them greater control of their movements by\u00a0optimizing the arrangement of sensors on their\u00a0bodies.<\/p>\n

\u201cYou can\u2019t put an infinite number of sensors on the robot itself,\u201d said<\/a> study co-lead author Andrew Spielberg, a PhD student at MIT CSAIL<\/a>. \u201cSo, the question is: how many sensors do you have, and where do you put those sensors in order to get the most bang for your buck?\u201d<\/p>\n

The researchers developed a novel neural network architecture to answer the question. It works by determining the ideal location for sensors and learning which motions are needed for different tasks.<\/p>\n

[Read: \u2018Sensorized\u2019 skin helps robots understand where the hell they are]<\/em><\/p>\n

The team first divided the robot\u2019s body into different regions, and used their rate of strain as inputs to the neural network. Over time, the network worked out the most efficient sequence of movements to complete tasks, such as which motions to use for gripping different options.<\/p>\n

The network also tracks which regions are used most frequently, and then removes the lesser-used areas from the inputs in subsequent trials.<\/p>\n

This process enables the system to recommend where to place sensors on the robot.\u00a0
\n