{"id":27535,"date":"2018-11-29T10:48:46","date_gmt":"2018-11-29T15:48:46","guid":{"rendered":"https:\/\/college.unc.edu\/?p=27535"},"modified":"2024-07-02T16:57:36","modified_gmt":"2024-07-02T16:57:36","slug":"ready-set-brake","status":"publish","type":"post","link":"https:\/\/collegearchive.unc.edu\/?p=27535","title":{"rendered":"Ready, Set, Brake"},"content":{"rendered":"<p><em>While autonomous vehicles begin to appear on roadways, gaps in knowledge are blocking the way to their full integration. Researchers at UNC are asking the tough questions to ensure that the driverless car picking you up will be safe for passengers, bicyclists, and pedestrians alike.<\/em><\/p>\n<figure id=\"attachment_27536\" aria-describedby=\"caption-attachment-27536\" style=\"width: 503px\" class=\"wp-caption aligncenter\"><img loading=\"lazy\" decoding=\"async\" class=\"wp-image-27536\" src=\"https:\/\/collegearchive.unc.edu\/wp-content\/uploads\/sites\/21\/2024\/07\/autonomous-cars-FA.jpg\" alt=\"Driverless cars graphic\" width=\"503\" height=\"335\" \/><figcaption id=\"caption-attachment-27536\" class=\"wp-caption-text\">How will autonomous vehicles be able to sense pedestrians and predict the path they will take? This is one of many questions that UNC computer scientists, city planners, and highway safety researchers are investigating with the new-age technology. Graphic by Cordina Cudebec.<\/figcaption><\/figure>\n<p>&nbsp;<\/p>\n<p>Just as no one could\u2019ve predicted the rise of social media in the last decade, no one today can predict precisely how driverless cars will look, operate, or shape our lives. All we know is that autonomous vehicles are coming. But they\u2019re coming slower than most people think.<\/p>\n<p>For every solution solved on the long road to ubiquitous self-driving cars, there are dozens of questions putting on the brakes. \u201cIt sounds like it\u2019s going to be this fantastic new world, but then you start picking apart all the different subtle questions in between,\u201d says <a href=\"http:\/\/www.hsrc.unc.edu\/team\/clamann\/\">Michael Clamann<\/a>, senior human factors engineer and autonomous vehicle expert with the <a href=\"http:\/\/www.hsrc.unc.edu\/\">UNC Highway Safety Research Center (HSRC)<\/a>.<\/p>\n<p>Those questions go beyond software and hardware. Making autonomous vehicles safe requires perfect alignment of computer science, engineering, psychology, sociology, and policy. The scope of societal impact goes further, reaching into a world that we can\u2019t yet imagine.<\/p>\n<p>\u201cWe\u2019re starting to recognize that technologies are complicated and thinking about the pluses and minuses is important,\u201d says <a href=\"https:\/\/planning.unc.edu\/faculty\/mcdonald\/\">Noreen McDonald<\/a>, chair of the Department of City and Regional Planning. \u201cIt\u2019s going to be a long time before our cities are a bunch of people driving around in autonomous vehicles.\u201d<\/p>\n<p>At UNC, researchers from a variety of disciplines are thinking through challenges from pedestrian safety to city parking, bringing us closer to this \u201cnew world\u201d one day at a time.<\/p>\n<p><strong>Building the brain<\/strong><\/p>\n<p>Autonomous vehicles use a variety of sensors to identify pedestrians and navigate safely. But going a step further and predicting their walking paths could reduce pedestrian accidents and fatalities even more, especially in urban areas. UNC computer scientist <a href=\"http:\/\/www.aniketbera.com\/\">Aniket Bera<\/a> and his team have developed one of the leading algorithms in the world for doing just that.<\/p>\n<p>\u201cFor autonomous vehicles, the biggest problems is how well they can learn and understand the surroundings,\u201d Bera says. \u201cIf it can learn the surroundings, it can drive efficiently and safely.\u201d<\/p>\n<p>Computer scientists have to train autonomous vehicles to identify individuals and groups as they stroll along crowded sidewalks using data-driven artificial intelligence and mathematical models.<\/p>\n<p>\u201cPeople tend to train models based on just data, but what we have done is try to make a combination of data and physically-based simulation,\u201d Bera says. This combination of psychology, artificial intelligence, simulation models, and mathematics has evolved through three stages over the years: pedestrian tracking, prediction, and behavior modeling.<\/p>\n<p>The first stage, pedestrian tracking, teaches the \u201cbrain\u201d of the autonomous vehicle the basics such as what a person looks like in a crowd where only a hand or shoulder may be visible in the camera sensors\u2019 live video feed. Once the computer learns that, it can predict where a pedestrian will go next. That\u2019s where pedestrian physics come into play.<\/p>\n<p>Bera and his colleagues discovered that as a crowd gets denser, its movement becomes similar to that of fluids. Using a combination of fluid dynamics and collision avoidance models, Bera\u2019s model can accurately predict where a pedestrian will walk several seconds into the future, with accuracy not dropping below 80 percent until after seven seconds. The system retrains itself with every new video frame, comparing the difference between its prediction and reality in real time.<\/p>\n<div class=\"ast-oembed-container \" style=\"height: 100%;\"><iframe loading=\"lazy\" title=\"Realtime Pedestrian Behavior Learning for Path Prediction and Navigation\" width=\"500\" height=\"281\" src=\"https:\/\/www.youtube.com\/embed\/37StrvOhDbs?feature=oembed\" frameborder=\"0\" allow=\"accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share\" referrerpolicy=\"strict-origin-when-cross-origin\" allowfullscreen><\/iframe><\/div>\n<p>Bera teamed up with the psychology department to take the algorithm yet another step forward to include behavior modeling. A data-driven process, thousands of simulations train the artificial intelligence to assign each pedestrian a combination of personality traits and behaviors, increasing the prediction\u2019s accuracy and influencing navigation decisions. For example, children typically walk more aggressively and are more likely to dart out into the road, so the car needs to slow down or stop in response.<\/p>\n<p>They\u2019re also trying to find cultural differences in walking styles. Pedestrians in western countries, for example, prefer more personal space between each other than people in eastern countries. \u201cThe system is agnostic to the culture, but it can also learn culture,\u201d Bera says. \u201cWe can train the algorithm to learn from hundreds of videos from multiple countries at the same time, and the artificial intelligence learns the patterns in just a few minutes. That would be almost impossible for humans.\u201d<\/p>\n<p>With Bera\u2019s research on human behavior modeling, he\u2019s showing future driverless cars how to see people as more than obstacles to avoid. \u201cA tree is different from a human being,\u201d Bera says. \u201cThere\u2019s emotion involved. There are ethics involved. We\u2019re trying to make autonomous vehicles smarter by giving them a brain that understands humans better \u2014 an effort I think few people in the world are working on.\u201d<\/p>\n<figure id=\"attachment_27538\" aria-describedby=\"caption-attachment-27538\" style=\"width: 505px\" class=\"wp-caption aligncenter\"><img loading=\"lazy\" decoding=\"async\" class=\" wp-image-27538\" src=\"https:\/\/collegearchive.unc.edu\/wp-content\/uploads\/sites\/21\/2024\/07\/ExpDemo-600x450-1.jpg\" alt=\"Driverless Cars Photo\" width=\"505\" height=\"379\" \/><figcaption id=\"caption-attachment-27538\" class=\"wp-caption-text\">To help driverless cars communicate with pedestrians, Clamann and his research team tested the effectiveness of signage on the front of the vehicle\u2019s front bumper.<\/figcaption><\/figure>\n<p><strong>Communicating with cars<\/strong><\/p>\n<p>While Bera focuses on the technology within autonomous vehicles, Clamann considers the situations in which humans and cars must interact.<\/p>\n<p>\u201cImagine that someone is driving you somewhere and you have a question. It\u2019s really easy to turn to the person and ask, <em>what\u2019s that<\/em> or <em>how much longer is it<\/em>,\u201d Clamann says. \u201cYou\u2019re still going to have those questions when you\u2019re using an autonomous vehicle.\u201d<\/p>\n<p>In his research, Clamann and his team test existing theories to determine the best practices for the self-driving car designers of the future, identifying the most efficient, safe, and inclusive ways for humans and said vehicles to communicate.<\/p>\n<p>Clamann specifically studies pedestrian-autonomous vehicle interactions. Today, when you approach a crosswalk, a car may slow down and the driver will wave you to cross from behind the wheel. But how do you communicate to pedestrians that it\u2019s safe to cross without a driver to wave them on?<\/p>\n<p>Leading ideas\u00a0include a display with text reading \u201csafe to cross.\u201d In transportation, all text is required to be readable from 100 feet away, translating to a 4-foot wide sign tacked onto the front of the car.<\/p>\n<p>\u201cNot to mention, if you\u2019re going to have a text-based message, what language are you going to have it in?\u201d Clamann says. \u201cA lot of people don\u2019t necessarily speak English, and we shouldn\u2019t exclude them from safety.\u201d<\/p>\n<p>Auto manufacturers have tried <a href=\"https:\/\/www.washingtonpost.com\/technology\/2018\/08\/29\/how-do-you-get-people-trust-autonomous-vehicles-this-company-is-giving-them-virtual-eyes\/?noredirect=on&amp;utm_term=.ff2d89233acd\">adding eyes<\/a> to the front of vehicles to simulate eye contact between pedestrians and drivers. This may work for one pedestrian at a time, but when there\u2019s a dense crowd, it\u2019s not possible to make eye contact with every person.<\/p>\n<p>\u201cWhat\u2019s most likely going to work is going to be very simple like a light, essentially the equivalent of a brake light, but on the front of the car,\u201d Clamann says. Using a light alerts pedestrians and gives information about the car\u2019s speed, rather than giving recommendations about what the pedestrian should do, eliminating problems that other methods create.<\/p>\n<p>To test these theories, Clamann and his team set up an experiment akin to a theatrical production on either a closed track or a road open to traffic, which can take weeks or months to plan.<\/p>\n<p>It took more than six months to set up a recent study to test the effectiveness of phone app that warns distracted pedestrians of oncoming traffic. First, they had to find the perfect location \u2014 somewhere with a blind corner so the car remains hidden from the pedestrian until the app\u2019s alert. After deciding on a closed track, they calculated and rehearsed the car\u2019s and pedestrian\u2019s exact route and timing and hired nearly a dozen people to execute the experiment\u2019s precise choreography.<\/p>\n<p>North Carolina is an ideal location for this research, as well as any other autonomous vehicle testing, as long as the car has a steering wheel and pedals. \u201cYou can test pretty much anywhere you want as long as it\u2019s done safely and in cooperation with the Department of Transportation, so we\u2019re open for doing many kinds of research efforts,\u201d Clamann says. \u201cI think that\u2019s not well understood by the research community here, so that\u2019s something that I\u2019m trying to help communicate.\u201d<\/p>\n<p><strong>Avoiding potholes<\/strong><\/p>\n<p>Within the Department of City and Regional Planning, McDonald and colleagues are asking the big questions about how autonomous vehicles will shape our communities. \u201cWe wanted to bring people together to think about the problems that these vehicles can solve, but also the problems they might create and how to mitigate those problems in advance,\u201d she says. Her team plans to examine a variety of impacts from shifting jobs to financing cities.<\/p>\n<p>Within HSRC, Clamann is gearing up to do the same for the relationship between driverless cars and schools. \u201cThere hasn\u2019t been a lot looking directly at how this is going to affect schools,\u201d Clamann says. \u201cWhether it\u2019s autonomous buses picking up kids, or what\u2019s going to happen when kids are being picked up and dropped off with autonomous vehicles, or the different regulations the counties have for vehicles going past schools and how\u2019s that going to change.\u2019\u2019<\/p>\n<p>There\u2019s also the issue of rural roads. Right now, self-driving car technology is dependent on using camera sensors to identify lane markings on well-mapped streets to stay on course. In rural areas, those markings start to disappear. Combined with other hurdles, such as low access to broadband, Wi-Fi, and electric charging stations, that means it\u2019ll be a while before autonomous vehicles make it into rural North Carolina.<\/p>\n<p>This slow rollout challenges the assumption that driverless cars will immediately make streets safer for pedestrians. Rural areas and spaces away from intersections \u2014 places where it\u2019s harder for autonomous vehicles to operate \u2014 won\u2019t benefit from this technology until much later.<\/p>\n<p>\u201cIf they were implemented ubiquitously, they could significantly reduce fatalities, but that\u2019s not reality,\u201d McDonald says. \u201cDesigning our communities to be safer is still important, because they\u2019re not going to solve the pedestrian fatalities problem tomorrow.\u201d<\/p>\n<p><strong>Identifying blind spots<\/strong><\/p>\n<p>Even with all of these problems solved, public opinion could still topple the implementation of autonomous vehicles.<\/p>\n<p>\u201cIf people are frightened of these things for whatever reason, they\u2019re not going to be accepted by the public, and then we\u2019re never going use them,\u201d Clamann says. \u201cThe majority of people are still scared,\u00a0so there\u2019s a lot of people trying to figure out what we can do to change that.\u201d<\/p>\n<p>After a 30 percent increase over a seven-year period, pedestrian fatalities reached the highest level in nearly three decades in 2016, according to the National Center for Statistics. Because driverless cars have the potential to drastically decrease those numbers, letting fear leave this technology on a test track may be a matter of life or death.<\/p>\n<p>Clamann and McDonald co-authored <a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0749379718320932?via%3Dihub\">a recent study on autonomous vehicles and pedestrian safety<\/a>, analyzing nearly 5,000 pedestrian deaths in 2016 to determine what percentage could have been prevented with driverless cars on the road. They found that there could be a significant reduction in crashes in certain conditions such as at dusk or dawn.<\/p>\n<p>\u201cAutonomous vehicles have better sensors than the human eye, so they essentially can see in the dark,\u201d McDonald says. \u201cThey should be better at detecting and eliminating that type of fatality or crash.\u201d<\/p>\n<p>&nbsp;<\/p>\n<figure id=\"attachment_27540\" aria-describedby=\"caption-attachment-27540\" style=\"width: 364px\" class=\"wp-caption aligncenter\"><img loading=\"lazy\" decoding=\"async\" class=\" wp-image-27540\" src=\"https:\/\/collegearchive.unc.edu\/wp-content\/uploads\/sites\/44\/2018\/11\/94-humanerror-347x500\" alt=\"National Motor Vehicle Crash Causation Survey graphic\" width=\"364\" height=\"525\" \/><figcaption id=\"caption-attachment-27540\" class=\"wp-caption-text\">These are the different ways in which human drivers become distracted on the road, according to the National Motor Vehicle Crash Causation Survey.<\/figcaption><\/figure>\n<p>&nbsp;<\/p>\n<p>But self-driving vehicles are not a panacea. Experts often cite the statistic that 94 percent of traffic fatalities are due to human error, a type of crash they say these vehicles can eliminate. If you break down that figure, Clamann argues, there\u2019s a different story.<\/p>\n<p>\u201cTwo percent of that 94 percent is people having heart attacks, which is considered a human error,\u201d Clamann says. \u201cThis whole idea in using autonomous vehicles and saying we\u2019re just going to eliminate human error needs to be looked at a little more carefully.\u201d<\/p>\n<p>The dozens of situations that constitute a human error \u2014 drunk driving, distracted driving, pedestrians darting into the road \u2014 all require different approaches, Clamann asserts. It\u2019ll take more time, cooperation, and research to determine the best way to design and implement this technology and steadily cut down that percentage.<\/p>\n<p>Outside of safety, it\u2019s too early to know what the fullest benefits of driverless cars are. Clamann predicts that we\u2019ll find out a decade or two after the initial shift to fully autonomous vehicles on the road.<\/p>\n<p>\u201cWhen is that? I have no idea,\u201d Clamann says. \u201cI\u2019m not sure it\u2019s going to happen in my lifetime, but I\u2019d love to see it.\u201d<\/p>\n<p><em>Michael Clamann is the senior human factors engineer and autonomous vehicle expert with the UNC Highway Safety Research Center.<\/em><\/p>\n<p><em>Noreen McDonald is the Thomas Willis Lambeth Distinguished Chair of Public Policy and chair of the Department of City and Regional Planning within the UNC College of Arts &amp; Sciences. She is also the director of the <a href=\"https:\/\/ctp.unc.edu\/\">Carolina Transportation Program<\/a>.<\/em><\/p>\n<p><em>Ankiet Bera is a research assistant professor in the Department of Computer Science within the College of Arts &amp; Sciences.<\/em><\/p>\n<p><em>Story by <a class=\"author url fn\" title=\"Posts by India Mackinson\" href=\"https:\/\/endeavors.unc.edu\/author\/indiamackinson\/\" rel=\"author\">India Mackinson<\/a>, <a href=\"https:\/\/endeavors.unc.edu\/ready-set-brake\/\">Endeavors<\/a><\/em><\/p>\n","protected":false},"excerpt":{"rendered":"<p>While autonomous vehicles begin to appear on roadways, gaps in knowledge are blocking the way to their full integration. Researchers at UNC are asking the tough questions to ensure that the driverless car picking you up will be safe for passengers, bicyclists, and pedestrians alike.<\/p>\n","protected":false},"author":4,"featured_media":27536,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_monsterinsights_skip_tracking":false,"_monsterinsights_sitenote_active":false,"_monsterinsights_sitenote_note":"","_monsterinsights_sitenote_category":0,"site-sidebar-layout":"default","site-content-layout":"","ast-site-content-layout":"default","site-content-style":"default","site-sidebar-style":"default","ast-global-header-display":"","ast-banner-title-visibility":"","ast-main-header-display":"","ast-hfb-above-header-display":"","ast-hfb-below-header-display":"","ast-hfb-mobile-header-display":"","site-post-title":"","ast-breadcrumbs-content":"","ast-featured-img":"","footer-sml-layout":"","ast-disable-related-posts":"","theme-transparent-header-meta":"","adv-header-id-meta":"","stick-header-meta":"","header-above-stick-meta":"","header-main-stick-meta":"","header-below-stick-meta":"","astra-migrate-meta-layouts":"default","ast-page-background-enabled":"default","ast-page-background-meta":{"desktop":{"background-color":"var(--ast-global-color-4)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"ast-content-background-meta":{"desktop":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"footnotes":""},"categories":[23,33,17,21,34,13],"tags":[24,514,25,26,74,55,36,38,39,40],"class_list":["post-27535","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-articles","category-media-news","category-natural-sciences-mathematics","category-news","category-news-archive","category-social-sciences","tag-carolina","tag-city-and-regional-planning","tag-college-of-arts-sciences","tag-college-of-arts-and-sciences","tag-computer-science","tag-research","tag-unc","tag-unc-college-of-arts-and-sciences","tag-unc-chapel-hill","tag-university-of-north-carolina-at-chapel-hill"],"_links":{"self":[{"href":"https:\/\/collegearchive.unc.edu\/index.php?rest_route=\/wp\/v2\/posts\/27535","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/collegearchive.unc.edu\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/collegearchive.unc.edu\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/collegearchive.unc.edu\/index.php?rest_route=\/wp\/v2\/users\/4"}],"replies":[{"embeddable":true,"href":"https:\/\/collegearchive.unc.edu\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=27535"}],"version-history":[{"count":2,"href":"https:\/\/collegearchive.unc.edu\/index.php?rest_route=\/wp\/v2\/posts\/27535\/revisions"}],"predecessor-version":[{"id":49018,"href":"https:\/\/collegearchive.unc.edu\/index.php?rest_route=\/wp\/v2\/posts\/27535\/revisions\/49018"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/collegearchive.unc.edu\/index.php?rest_route=\/wp\/v2\/media\/27536"}],"wp:attachment":[{"href":"https:\/\/collegearchive.unc.edu\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=27535"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/collegearchive.unc.edu\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=27535"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/collegearchive.unc.edu\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=27535"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}