Tag Archives: automotive

Discussion on self-driving cars at Science Cafe Concord

This week I had a chance to participate in a panel discussion on self-driving cars at the Concord Science Cafe, organized by David Brooks. David is a science writer for the Concord Monitor – thanks David for inviting me! The other panelists were Joe Cunningham of NHTI, and Sean Smith of Dartmouth. Joe’s background is in robotics and automation, while Sean is a computer scientist with a focus on information security.

Science Cafe brings together experts from a particular field and an audience made up of members of the community who are interested in that topic. The audience brings up questions and problems, and the panel tries to provide insight from their own perspective. In our case this resulted in a really fun exchange of ideas between David who was moderating, the panelists, and the audience of about 50.

Some of the things I learned from this panel:

  • You should organize an event like this at a bar, in order to make everyone relax and feel comfortable.
  • Many people worry about the capabilities of self-driving cars: are they really better than humans?
  • Tire pressure sensors can be a security risk because they can be detected from outside the vehicle, even to set off an explosive meant for the passengers of that vehicle.
  • Owners of self-driving cars might look for modifications to the car’s software, e.g. to set up their vehicle to exceed the speed limit. Thus, even with self-driving cars we will still need traffic rule enforcement of some kind.

 

UNH IRES: HCI summer student research experience in Germany

HCI Lab, Stuttgart

UNH ECE professor Tom Miller and I were recently awarded an NSF International Research Experiences for Students (IRES) grant. Our IRES grant will fund students conducting research at the University of Stuttgart in Germany.

Albrecht Schmidt

Under our NSF IRES grant, each summer between 2014 and 2017, three undergraduate and three graduate students  will conduct research for just under 9 weeks at the Human Computer Interaction (HCI) Lab of Professor Albrecht Schmidt at the University of Stuttgart. Professor Schmidt and his lab are among the world leaders in the field of HCI.

Student research will focus on two areas: in-vehicle speech interaction and speech interaction with public displays. For in-vehicle speech, students will relate the benefits and limitations of speech interaction with in-vehicle devices with real-world parameters, such as how well speech recognition works at any given moment. They will also work to identify why it is that talking to a passenger appears to reduce the probability of a crash, and how we might be able to use this new information to create safer in-vehicle speech interactions. Similarly, students will explore how speech interaction can allow smooth interaction with electronic public displays.

Stuttgart Palace Square (Stefan Fussan: https://www.flickr.com/photos/derfussi/)

Successful applicants will receive full financial support for participation, covering items such as airfare, room and board, health insurance, as well as a $500/week stipend. The total value of the financial package is approximately $8,500 for 9 weeks.

Details about the program, including applications instructions, are available here. Please note that this program is only available to US citizens and permanent residents. If you have questions please contact Andrew Kun (andrew dot kun at unh dot edu) or Tom Miller (tom dot miller at unh dot edu).

2014 visit to University College London

Last week I travelled to London to give a talk at University College London (UCL). My host was Duncan Brumby, who also recently visited us at UNH. My talk introduced our work on in-vehicle human-computer interaction, touching on subjects from Project54 to driving simulator-based experiments.

It was great to talk to Duncan again, and it was really nice to meet some of his colleagues, including Anna Cox, Paul Marshall, and Sandy Gould. Thanks to all for hosting me.

While my trip was brief, I did get a chance to also visit the British Museum. This is one of my favorite places in the world, and here’s a photo of my favorite artifact from the museum’s vast collection, the Rosetta Stone:

You can see more pictures from my trip on Flickr.

 

 

 

2014 Visit to University of Szeged

The home of the Research Group on Artificial Intelligence

Yesterday I visited the University of Szeged. I was there at the invitation of fellow Fulbrighter Mark Jelasity. Mark is a member of the Research Group on Artificial Intelligence. He and I met at a recent event organized by Fulbright Hungary, where Hungarian Fulbright scholars reported on their experiences in the USA.

The primary purpose of my visit was to give a talk at the Institute of Informatics; thanks to Peter Szabó for including my talk in the Institute’s seminar series. I was really excited, as this was the first time in my career that I gave a talk in Hungarian. For fun, here are the Hungarian title and abstract:

Szemkövetők használata autóvezetéses szimulációs kisérletekben

Absztrakt: A University of New Hampshire kutatói több mint egy évtizede foglalkoznak a járműveken belüli ember-gép interfészekkel. Ez az előadás először egy rövid áttekintést nyújt a rendőr járművekre tervezett Project54 rendszer fejlesztéséről és telepítéséről. A rendszer különböző modalitású felhasználói felületeket biztosít, beleértve a beszéd modalitást. A továbbiakban az előadás beszámol közelmúltban végzett autóvezetés-szimulációs kísérletekről, amelyekben a szimulátor és egy szemkövető adatai alapján becsültük a vezető kognitív terhelését, vezetési teljesítményét, és vizuális figyelmét a külső világra.

As part of the visit I had a chance to talk to Mark about his research (see Mark’s Google Scholar profile). Mark’s interest is in distributed learning, which might have fascinating applications in the automotive domain.

Szeged fish soup

My visit was also exciting on a personal level, as I was born in Szeged, and my family often travelled there during my childhood. I walked around the town a bit to reminisce, and Mark treated me to an excellent lunch of fish soup.

See pictures from my trip on Flickr.

2013 Liberty Mutual visit to UNH ECE

Yesterday I hosted four researchers from the Liberty Mutual Research Institute for Safety: Bill Horrey, Yulan Liang, Angela Garabet, and Luci Simmons. This visit follows my recent visit to Liberty Mutual this summer.

As part of the visit, Bill gave a talk to my ECE 900 class. He discussed the wide variety of research performed at his institute, with an emphasis on the vehicle-related work that he is involved in. As part of this work Bill and colleagues conduct studies on a test track with an instrumented vehicle, which they brought along:

After the talk Tom Miller and I had a chance to show our visitors our driving simulator lab and discuss a host of research issues. It was fun – thanks Bill, Yulan, Angela and Luci.

See pictures from the visit on Flickr.

 

Impressions from the 2013 FHWA workshop on utilizing various data sources for research

Last week I participated in the work of a workshop organized by the FHWA which explored utilizing various data sources for surface transportation human factors research. Here are some impressions from the workshop.

Bicycles are cool
Marco Dozza explores the use of bicycles and has a fleet of instrumented bikes. I really liked the elegance of his approach.

We can use technology to warn pedestrians (even when they have headphones)
Toru Hagiwara and Hidekatsu Hamaoka discussed research on protecting pedestrians in crosswalks. I can relate to this: ever since I started walking with kids, I’m sorely aware of the dangers of crosswalks.

We need a variety of data sources
Michael Manser, Sue Chrysler, John Lee and Linda Boyle gave presentations on a variety of data sources they’ve used: from laboratory studies to naturalistic driving data. They all agree that there’s a need to combine results from different data sources in order to find solutions to human factors problems. I was really impressed with John Lee’s discussion of the use of Twitter to understand traffic.

Panel discussions can be informative…
… when you have a skilled moderator (Don Fisher) and engaged participants (the presenters listed above). Don’s theme for the discussion (the 3 Cs): our data should be comprehensive and complementary, but it is sometimes also contradictory. The discussion brought up ideas such as:

  • there’s a need for standardization (see Paul Green‘s work on SAE J2944 [1]);
  • Linda Boyle points out that there’s a fourth C: confusion;
  • Sue Chrysler points out that discussions about data often focus on the “how?” (how can we collect, process, interpret data?). But we need to first resolve the “why?” and “what?” questions. I really appreciated this comment, as it is the central point I make in my ECE 900 course.

Thanks to FHWA’s David Yang for hosting the workshop – it was informative and fun. See my pictures from the workshop on Flickr.

References

[1] Paul Green, “Standard Definitions for Driving Measures and Statistics: Overview and Status of Recommended Practice J2944,” AutomotiveUI 2013

 

AutomotiveUI 2013: A brief report

At the end of October I was in Eindhoven for AutomotiveUI 2013. The conference was hosted by Jacques Terken, who along with his team did a splendid job: from the technical content, to the venue, to the banquet, and the invited demonstrations, everything was of high quality and ran smoothly. Thanks Jacques and colleagues!

The conference started with workshops, including our CLW 2013. The workshop was sponsored by Microsoft Research and it brought together over 30 participants. In addition to six contributed talks, the workshop featured two invited speakers. The keynote lecture was given by Klaus Bengler who discussed challenges with cooperative driving. Tuhin Diptiman discussed the implications of the 2013 NHTSA visual manual guidelines on the design of in-vehicle interfaces. Thanks Klaus and Tuhin for your engaging presentations! See pictures from CLW on Flickr.

The main conference included talks, posters and demos. Our group was productive: we had one talk and three posters. One of the memorable moments of the conference was the invited demos. I had a chance to ride in a TNO car using Cooperative Adaptive Cruise Control (CACC). See the video below (the driver and narrator is TNO researcher E. (Ellen) van Nunen – thanks Ellen!

See my pictures from the conference on Flickr.

Visit to Liberty Mutual Research Institute for Safety

Last month I had a chance to visit the Liberty Mutual Research Institute for Safety in Hopkinton, MA. My host was Bill Horrey, senior research scientist at the institute. I also had a chance to talk to Marvin Dainoff, Yulan Liang, Angela Garabet, and a number of other researchers.

The institute is truly impressive. First, as Marvin explained to me, this is an organization devoted to independent research. The institute is funded by Liberty Mutual, but sets its own agenda. The only “constraint” on this agenda is that the research has to be related to Liberty Mutual’s business. Given that this business covers areas from driving, to homes, to health care, this is hardly a constraint. Furthermore, the institute is committed to publishing its work in peer-reviewed publications. In fact, publications are the institute’s central measure of success. The institute has a hallway with three (!) whiteboards, where staff keep track of their publications for the year. Each year the goal is to fill all three boards by December.

The institute has a number of very interesting labs. Of course for me the most interesting one was the driving simulator lab (the instrumented vehicle is a very close second!). The simulator is made by Real Time Technologies and the lab also has a head-mounted eye tracker.

So thanks to Bill and colleagues for hosting me. I really enjoyed talking to them about some research problems (including their recent work on drowsy drivers [1]), as well as the technical details of running a simulator lab. You can see a few more pictures about my visit on Flickr.

 

References

[1] William J. Horrey, Yulan Liang, Michael L. Lee, Mark E. Howard, Clare Anderson, Michael S. Shreeve, Conor O’Brien, Charles A. Czeisler, “The Long Road Home: Driving Performance and Ocular Measurements of Drowsiness Following Night Shift-Work,” Driving Assessment 2013

Lee Slezak visit

At the end of March we hosted Lee Slezak in our lab. Lee is Vehicle Systems and Testing Manager in DOE’s Vehicle Technology Program.

As part of the visit Lee saw our driving simulator lab, and we had a chance to discuss with him a number of projects, including work on a UNH ECE senior project that deals with personal navigation devices. Following this, Lee introduced us to the work of DOE’s Vehicle Technology Program. Three points from the presentation really resonated with me:

  1. Interoperability is a problem. As we have found in our work on in-vehicle devices for police cruisers (Project54) [1], and as many UNH ECE students find in working at UNH’s IOL, many devices are not designed with interoperability in mind. This is also true of the different devices people use to charge batteries in vehicles.
  2. Simulation can help focus R&D efforts. Our work on augmented reality personal navigation devices is a good example of using (driving) simulation to explore the value of an idea that is not yet technologically feasible (at least not with a reasonable price tag) [2]. DOE explores a number of technologies in simulation before committing to developing and testing hardware.
  3. Green racing can move automotive technology forward. Innovations used in car racing used to be transferable to consumer vehicles. This is almost never the case any more. Green racing is trying to change this: the green technologies that can win races can also provide improvements for everyday vehicles.

Here’s a photo of Lee discussing his ideas with us in the lab. For more photos see Flickr.

So, thanks Lee for taking the time to visit – it was a valuable and it was fun. I would also like to thank Kristin Bennett and Jan Nisbet for making Lee’s visit possible.

 

References

[1] Andrew L. Kun, W. Thomas Miller, III and William H. Lenharth, “Computers in police cruisers,” IEEE Pervasive Computing, Vol. 3, Issue 4, pp.: 34 – 41, Oct.-Dec. 2004

[2] Zeljko Medenica, Andrew L. Kun, Tim Paek, Oskar Palinko, “Augmented Reality vs. Street Views: A Driving Simulator Study Comparing Two Emerging Navigation Aids,” MobileHCI 2011

New York Times article discusses our work on in-vehicle navigation devices

Last week I was interviewed by Randall Stross for an article that appeared in the September 2 edition of the New York Times. Mr. Stross’ article, “When GPS Confuses, You May Be to Blame,” discusses research on in-vehicle personal navigation devices, including our work on comparing voice-only instructions to map+voice instructions [1].

Specifically, Mr. Stross reports on a driving simulator study published at AutomotiveUI 2009, in which we found that drivers spent significantly more time looking at the road ahead when navigation instructions were provided using a voice-only interface than in the case when both voice instructions and a map were available. In fact, with voice-only instructions drivers spent about 4 seconds more every minute looking at the road ahead. Furthermore, we found evidence that this difference in the time spent looking at the road ahead also had an effect on driving performance measures. These results led us to conclude that voice-only instructions might be safer to use than voice+map instructions. However, the majority of our participants preferred having a map in addition to the voice instructions.

This latter finding was the impetus for a follow-on study in which we explored projecting navigation instructions onto the real world scene (using augmented reality) [2]. We found that augmented reality navigation aids allow for excellent visual attention to the road ahead and excellent driving performance.

References

[1] Andrew L. Kun, Tim Paek, Zeljko Medenica, Nemanja Memarovic, Oskar Palinko, “Glancing at Personal Navigation Devices Can Affect Driving: Experimental Results and Design Implications,” Automotive UI 2009

[2] Zeljko Medenica, Andrew L. Kun, Tim Paek, Oskar Palinko, “Augmented Reality vs. Street Views: A Driving Simulator Study Comparing Two Emerging Navigation Aids,” MobileHCI 2011