{"id":4641,"date":"2016-02-17T10:00:45","date_gmt":"2016-02-17T15:00:45","guid":{"rendered":"https:\/\/www.bu.edu\/federal\/?p=4641"},"modified":"2016-02-19T11:44:03","modified_gmt":"2016-02-19T16:44:03","slug":"urop-students-project-a-thinking-robot","status":"publish","type":"post","link":"https:\/\/www.bu.edu\/federal\/2016\/02\/17\/urop-students-project-a-thinking-robot\/","title":{"rendered":"UROP Student\u2019s Project: A Thinking Robot"},"content":{"rendered":"

Self-directed \u2018bot can identify objects<\/h2>\n
\n

In the following video,<\/a><\/strong> watch Emily Fitzgerald\u2019s artificially intelligent robot.<\/p>\n<\/div>\n

\u201cThat is a ball.\u201d \u201cI do believe that is a cone.\u201d \u201cSeems like a wonderful book.\u201d <\/em>The voice is mechanical and flat, and anyone offering such banal commentary and sounding so bored would surely bomb in a job interview. But in this case, the observations are impressive. They\u2019re made by what looks like a two-foot-tall stack of hors d\u2019oeuvre trays on wheels, careening around the floor and proclaiming its discoveries as its \u201ceye,\u201d an attached camera, falls on them.<\/p>\n

This robot has learned to recognize these specific objects\u2014and to steer around obstacles, albeit clumsily\u2014without human guidance. Its camera sends information about what it sees to a laptop sitting atop the robot; the laptop in turn communicates with a laboratory desktop, whose monitor flashes whatever the robot\u2019s camera catches.<\/p>\n

\u201cIt\u2019s almost self-thinking\u201d in its ability to get around roadblocks, says Emily Fitzgerald (ENG\u201916), who bestowed the \u2019bot with a brain as her project last summer with the Undergraduate Research Opportunities Program<\/a>\u00a0(UROP), which provides funding for faculty-mentored research by undergrad students. More important than the robot\u2019s autonomous navigation, she says, is its ability to recognize specific objects.<\/p>\n

Such self-guiding, object-spotting robots<\/a> are a Holy Grail for scientists, with potential applications that include exploring distant planets\u2019 landscapes. In Fitzgerald\u2019s case, she used a deep neural network<\/a>, a form of artificial intelligence that simulates brain neurons. Deep neural networks process huge amounts of data to solve problems, like recognizing a ball or cone.<\/p>\n

\u201cThere\u2019s an algorithm that will take a ton of pictures of one object and will put it in and compile it all,\u201d says Fitzgerald. \u201cThen we basically assign a number to it.\u201d The robot \u201cwill come upon an object and it will say, \u2018Oh, there\u2019s an object in front of me, let me think about it.\u2019 It will\u2026find a picture that corresponds with the object, pick that number, and then it will be able to use that as a reference, so it can exclaim, \u2018Oh, it\u2019s a ball,\u2019 \u2018It\u2019s a cone,\u2019 or whatever object I had decided to teach it.\u201d<\/p>\n

Massimiliano Versace (GRS\u201907), a College of Arts & Sciences research assistant professor and director of BU\u2019s Neuromorphics Lab<\/a>, oversaw Fitzgerald\u2019s UROP project, and she had help from Lucas Neves (ENG\u201916), a volunteer in Versace\u2019s lab, and Matthew Luciw, a visiting researcher at BU\u2019s Center for Computational Neuroscience and Neural Technology<\/a>.<\/p>\n

Asked how hard it was to train their metallic pupil in object recognition, the team members laugh. \u201cThere were quite a few times where we did despair a little bit that, you know, this wasn\u2019t going to work,\u201d says Fitzgerald, who first had to master an unfamiliar programming language. Then the team needed to make sure that the array of different software in the project would work together \u201cwithout crashing the system,\u201d she says.<\/p>\n

Often, the software wasn\u2019t compatible, resulting in a somewhat ditsy robot. \u201cMost of the time, it just didn\u2019t start,\u201d Neves says, ruefully recalling those tough moments. It also could get lost: sensors in its wheels tell the robot how far it\u2019s traveled. But \u201cthe wheels weren\u2019t moving at a constant rate, so whenever the robot would shoot off, it would think it had gone farther than it had because the wheels spun faster,\u201d says Fitzgerald.<\/p>\n

So the Terminator it isn\u2019t. Whether Fitzgerald\u2019s project will yield a commercial application someday remains an open question, says Versace, but he has no doubt about the viability of this type of work. Versace heads Neurala<\/a>, a BU spin-off company, and members of his lab met recently with NASA to discuss related research.<\/p>\n

As for Fitzgerald, who was turned on to engineering after excelling at physics and math in high school, she says the project persuaded her to pursue a career in bioimaging. Someday, she says, robotic surgical devices running off neural networks will detect objects in human patients.<\/p>\n

\u201cI\u2019ve actually taken this project and I\u2019ve said, OK, what else can I do with it in the biomedical setting as well?\u201d she says. \u201cIt\u2019s really shaped how I\u2019ve thought about my future going forward.\u201d<\/p>\n

Author, Rich Barlow can be reached at barlowr@bu.edu<\/a>.<\/em><\/p>\n","protected":false},"excerpt":{"rendered":"

Self-directed \u2018bot can identify objects In the following video, watch Emily Fitzgerald\u2019s artificially intelligent robot. \u201cThat is a ball.\u201d \u201cI do believe that is a cone.\u201d \u201cSeems like a wonderful book.\u201d The voice is mechanical and flat, and anyone offering such banal commentary and sounding so bored would surely bomb in a job interview. But […]<\/p>\n","protected":false},"author":7048,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":[],"categories":[8],"tags":[34,88,109,142,67,41,13,93,5,220,65,39,66,221],"_links":{"self":[{"href":"https:\/\/www.bu.edu\/federal\/wp-json\/wp\/v2\/posts\/4641"}],"collection":[{"href":"https:\/\/www.bu.edu\/federal\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.bu.edu\/federal\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.bu.edu\/federal\/wp-json\/wp\/v2\/users\/7048"}],"replies":[{"embeddable":true,"href":"https:\/\/www.bu.edu\/federal\/wp-json\/wp\/v2\/comments?post=4641"}],"version-history":[{"count":2,"href":"https:\/\/www.bu.edu\/federal\/wp-json\/wp\/v2\/posts\/4641\/revisions"}],"predecessor-version":[{"id":4643,"href":"https:\/\/www.bu.edu\/federal\/wp-json\/wp\/v2\/posts\/4641\/revisions\/4643"}],"wp:attachment":[{"href":"https:\/\/www.bu.edu\/federal\/wp-json\/wp\/v2\/media?parent=4641"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.bu.edu\/federal\/wp-json\/wp\/v2\/categories?post=4641"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.bu.edu\/federal\/wp-json\/wp\/v2\/tags?post=4641"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}