{"id":19702,"date":"2021-01-22T21:01:49","date_gmt":"2021-01-22T15:16:49","guid":{"rendered":"https:\/\/www.revoscience.com\/en\/?p=19702"},"modified":"2021-01-22T21:02:23","modified_gmt":"2021-01-22T15:17:23","slug":"designing-customized-brains-for-robots","status":"publish","type":"post","link":"https:\/\/www.revoscience.com\/en\/designing-customized-brains-for-robots\/","title":{"rendered":"Designing customized \u201cbrains\u201d for robots"},"content":{"rendered":"\n
\"\"<\/figure>\n\n\n\n

CAMBRIDGE, Mass. (MIT News)–\u00a0Contemporary robots\u00a0can<\/em>\u00a0move quickly. \u201cThe motors are fast, and they\u2019re powerful,\u201d says Sabrina Neuman.<\/p>\n\n\n\n

Yet in complex situations, like interactions with people, robots often don\u2019t<\/em> move quickly. \u201cThe hang up is what\u2019s going on in the robot\u2019s head,\u201d she adds.<\/p>\n\n\n\n

Perceiving stimuli and calculating a response takes a \u201cboatload of computation,\u201d which limits reaction time, says Neuman, who recently graduated with a PhD from the MIT Computer Science and Artificial Intelligence Laboratory (CSAIL). Neuman has found a way to fight this mismatch between a robot\u2019s \u201cmind\u201d and body. The method, called robomorphic computing, uses a robot\u2019s physical layout and intended applications to generate a customized computer chip that minimizes the robot\u2019s response time.<\/p>\n\n\n\n

The advance could fuel a variety of robotics applications, including, potentially, frontline medical care of contagious patients. \u201cIt would be fantastic if we could have robots that could help reduce risk for patients and hospital workers,\u201d says Neuman.<\/p>\n\n\n\n

Neuman will present the research at this April\u2019s International Conference on Architectural Support for Programming Languages and Operating Systems. MIT co-authors include graduate student Thomas Bourgeat and Srini Devadas, the Edwin Sibley Webster Professor of Electrical Engineering and Neuman\u2019s PhD advisor. Other co-authors include Brian Plancher, Thierry Tambe, and Vijay Janapa Reddi, all of Harvard University. Neuman is now a postdoctoral NSF Computing Innovation Fellow at Harvard\u2019s School of Engineering and Applied Sciences.<\/p>\n\n\n\n

There are three main steps in a robot\u2019s operation, according to Neuman. The first is perception, which includes gathering data using sensors or cameras. The second is mapping and localization: \u201cBased on what they\u2019ve seen, they have to construct a map of the world around them and then localize themselves within that map,\u201d says Neuman. The third step is motion planning and control \u2014 in other words, plotting a course of action.<\/p>\n\n\n\n

These steps can take time and an awful lot of computing power. \u201cFor robots to be deployed into the field and safely operate in dynamic environments around humans, they need to be able to think and react very quickly,\u201d says Plancher. \u201cCurrent algorithms cannot be run on current CPU hardware fast enough.\u201d<\/p>\n\n\n\n

Neuman adds that researchers have been investigating better algorithms, but she thinks software improvements alone aren\u2019t the answer. \u201cWhat\u2019s relatively new is the idea that you might also explore better hardware.\u201d That means moving beyond a standard-issue CPU processing chip that comprises a robot\u2019s brain \u2014 with the help of hardware acceleration.<\/p>\n\n\n\n

Hardware acceleration refers to the use of a specialized hardware unit to perform certain computing tasks more efficiently. A commonly used hardware accelerator is the graphics processing unit (GPU), a chip specialized for parallel processing. These devices are handy for graphics because their parallel structure allows them to simultaneously process thousands of pixels. \u201cA GPU is not the best at everything, but it\u2019s the best at what it\u2019s built for,\u201d says Neuman. \u201cYou get higher performance for a particular application.\u201d Most robots are designed with an intended set of applications and could therefore benefit from hardware acceleration. That\u2019s why Neuman\u2019s team developed robomorphic computing.<\/p>\n\n\n\n

The system creates a customized hardware design to best serve a particular robot\u2019s computing needs. The user inputs the parameters of a robot, like its limb layout and how its various joints can move. Neuman\u2019s system translates these physical properties into mathematical matrices. These matrices are \u201csparse,\u201d meaning they contain many zero values that roughly correspond to movements that are impossible given a robot\u2019s particular anatomy. (Similarly, your arm\u2019s movements are limited because it can only bend at certain joints \u2014 it\u2019s not an infinitely pliable spaghetti noodle.)<\/p>\n\n\n\n

The system then designs a hardware architecture specialized to run calculations only on the non-zero values in the matrices. The resulting chip design is therefore tailored to maximize efficiency for the robot\u2019s computing needs. And that customization paid off in testing.<\/p>\n\n\n\n

Hardware architecture designed using this method for a particular application outperformed off-the-shelf CPU and GPU units. While Neuman\u2019s team didn\u2019t fabricate a specialized chip from scratch, they programmed a customizable field-programmable gate array (FPGA) chip according to their system\u2019s suggestions. Despite operating at a slower clock rate, that chip performed eight times faster than the CPU and 86 times faster than the GPU.<\/p>\n\n\n\n

\u201cI was thrilled with those results,\u201d says Neuman. \u201cEven though we were hamstrung by the lower clock speed, we made up for it by just being more efficient.\u201d<\/p>\n\n\n\n

Plancher sees widespread potential for robomorphic computing. \u201cIdeally we can eventually fabricate a custom motion-planning chip for every robot, allowing them to quickly compute safe and efficient motions,\u201d he says. \u201cI wouldn’t be surprised if 20 years from now every robot had a handful of custom computer chips powering it, and this could be one of them.\u201d Neuman adds that robomorphic computing might allow robots to relieve humans of risk in a range of settings, such as caring for covid-19 patients or manipulating heavy objects.<\/p>\n\n\n\n

Neuman next plans to automate the entire system of robomorphic computing. Users will simply drag and drop their robot\u2019s parameters, and \u201cout the other end comes the hardware description. I think that\u2019s the thing that\u2019ll push it over the edge and make it really useful.\u201d<\/p>\n","protected":false},"excerpt":{"rendered":"

Contemporary robots can move quickly. \u201cThe motors are fast, and they\u2019re powerful,\u201d says Sabrina Neuman.<\/p>\n","protected":false},"author":2,"featured_media":19703,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[14,47],"tags":[],"class_list":["post-19702","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-innovation","category-it"],"featured_image_urls":{"full":["https:\/\/www.revoscience.com\/en\/wp-content\/uploads\/2021\/01\/Robomorphic-01-press.jpg",900,600,false],"thumbnail":["https:\/\/www.revoscience.com\/en\/wp-content\/uploads\/2021\/01\/Robomorphic-01-press-200x200.jpg",200,200,true],"medium":["https:\/\/www.revoscience.com\/en\/wp-content\/uploads\/2021\/01\/Robomorphic-01-press-600x400.jpg",600,400,true],"medium_large":["https:\/\/www.revoscience.com\/en\/wp-content\/uploads\/2021\/01\/Robomorphic-01-press-768x512.jpg",750,500,true],"large":["https:\/\/www.revoscience.com\/en\/wp-content\/uploads\/2021\/01\/Robomorphic-01-press-675x450.jpg",675,450,true],"1536x1536":["https:\/\/www.revoscience.com\/en\/wp-content\/uploads\/2021\/01\/Robomorphic-01-press.jpg",900,600,false],"2048x2048":["https:\/\/www.revoscience.com\/en\/wp-content\/uploads\/2021\/01\/Robomorphic-01-press.jpg",900,600,false],"ultp_layout_landscape_large":["https:\/\/www.revoscience.com\/en\/wp-content\/uploads\/2021\/01\/Robomorphic-01-press.jpg",900,600,false],"ultp_layout_landscape":["https:\/\/www.revoscience.com\/en\/wp-content\/uploads\/2021\/01\/Robomorphic-01-press.jpg",855,570,false],"ultp_layout_portrait":["https:\/\/www.revoscience.com\/en\/wp-content\/uploads\/2021\/01\/Robomorphic-01-press.jpg",600,400,false],"ultp_layout_square":["https:\/\/www.revoscience.com\/en\/wp-content\/uploads\/2021\/01\/Robomorphic-01-press.jpg",600,400,false],"newspaper-x-single-post":["https:\/\/www.revoscience.com\/en\/wp-content\/uploads\/2021\/01\/Robomorphic-01-press-760x490.jpg",760,490,true],"newspaper-x-recent-post-big":["https:\/\/www.revoscience.com\/en\/wp-content\/uploads\/2021\/01\/Robomorphic-01-press-550x360.jpg",550,360,true],"newspaper-x-recent-post-list-image":["https:\/\/www.revoscience.com\/en\/wp-content\/uploads\/2021\/01\/Robomorphic-01-press-95x65.jpg",95,65,true],"web-stories-poster-portrait":["https:\/\/www.revoscience.com\/en\/wp-content\/uploads\/2021\/01\/Robomorphic-01-press.jpg",640,427,false],"web-stories-publisher-logo":["https:\/\/www.revoscience.com\/en\/wp-content\/uploads\/2021\/01\/Robomorphic-01-press.jpg",96,64,false],"web-stories-thumbnail":["https:\/\/www.revoscience.com\/en\/wp-content\/uploads\/2021\/01\/Robomorphic-01-press.jpg",150,100,false]},"author_info":{"info":["RevoScience"]},"category_info":"Innovation<\/a> IT<\/a>","tag_info":"IT","comment_count":"0","_links":{"self":[{"href":"https:\/\/www.revoscience.com\/en\/wp-json\/wp\/v2\/posts\/19702","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.revoscience.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.revoscience.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.revoscience.com\/en\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.revoscience.com\/en\/wp-json\/wp\/v2\/comments?post=19702"}],"version-history":[{"count":0,"href":"https:\/\/www.revoscience.com\/en\/wp-json\/wp\/v2\/posts\/19702\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.revoscience.com\/en\/wp-json\/wp\/v2\/media\/19703"}],"wp:attachment":[{"href":"https:\/\/www.revoscience.com\/en\/wp-json\/wp\/v2\/media?parent=19702"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.revoscience.com\/en\/wp-json\/wp\/v2\/categories?post=19702"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.revoscience.com\/en\/wp-json\/wp\/v2\/tags?post=19702"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}