Intelligent Transportation Systems

Friday Seminar: Road Vehicle Automation History, Opportunities, and Challenges

Automated Car

It's a new year, a new semester, and a new TRANSOC Friday Seminar! This week PATH researcher Steven Shladover presents, "Road Vehicle Automation History, Opportunities, and Challenges".

Road vehicle automation has recently attracted intense interest from the media, the general public and now the transportation community. This interest is largely based on serious misconceptions about the level of automation of road vehicles that is likely to be achievable within the foreseeable future. This presentation addresses those misconceptions, beginning with a historical overview going back to 1939, and continuing with definition of multiple levels of vehicle automation. The importance of communication and cooperation among automated vehicles and between these vehicles and the roadway infrastructure is illustrated with examples from experiments conducted at the PATH Program. The technical challenges that remain to be resolved before fully automated driving can become reality are explained.

The seminar will take place Friday, January 31, 2014 from 4 - 5 p.m. in 534 Davis Hall. And of course, Cookie Hour returns preceding the Seminar at 3:30 in the Library. See you then!

Uber to expand into "urban logistics"?

Travis Kalanick, Co-Founder & CEO, Uber @ LeWeb Paris Day 1 2013-2555

Transportation Network Company, or as most people refer to them, rideshare company Uber is looking to expand its market. This week Chief Executive Travis Kalanick spoke a Le Web, where he described the Uber's plans to enter "urban logistics". He said, "Today, we are in the business of delivering cars in five minutes. Once you're in the business of delivering cars in five minutes, there are a lot of things you can deliver in five minutes."

Phase 2 will build upon notable promotions as delivering kittens, ice cream, and Christmas trees

Transportation mode recognition using GPS and accelerometer data

Cyclists

One of the big problems for smartphone travel diary apps is automatic mode detection. The split between walking and not is pretty easy, as is cycling, but what about separating cars from rail? Apps like Moves just dubs it "transport", but that doesn't help much with travel behavior research. A new paper in Transportarion Researc Part C examines using accelerometers and GPS to detect mode. Tao Feng and Harry J.P. Timmermans from Eindhoven University of Technology present their research in, "Transportation mode recognition using GPS and accelerometer data"

Potential advantages of global positioning systems (GPS) in collecting travel behavior data have been discussed in several publications and evidenced in many recent studies. Most applications depend on GPS information only. However, transportation mode detection that relies only on GPS information may be erroneous due to variance in device performance and settings, and the environment in which measurements are made. Accelerometers, being used mainly for identifying peoples’ physical activities, may offer new opportunities as these devices record data independent of exterior contexts. The purpose of this paper is therefore to examine the merits of employing accelerometer data in combination with GPS data in transportation mode identification. Three approaches (GPS data only, accelerometer data only and a combination of both accelerometer and GPS data) are examined. A Bayesian Belief Network model is used to infer transportation modes and activity episodes simultaneously. Results show that the use of accelerometer data can make a substantial contribution to successful imputation of transportation mode. The accelerometer only approach outperforms the GPS only approach in terms of the predictive accuracy. The approach which combines GPS and accelerometer data yields the best performance.

The full article can be found here

Friday Seminar: Decomposing connected vehicle dynamics: delay effects and nonlinearities

Interior: 2013 SRT Viper & SRT Viper GTS

Today's TRANSOC Friday seminar is about connected vehicles. University of Michigan's Gabor Orosz will present, "Decomposing connected vehicle dynamics: delay effects and nonlinearities".

Arising technologies related to vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communications can significantly improve the efficiency of connected vehicle systems. These allow cars to obtain detailed information about the motion of distant vehicles. Such information can be presented to the driver or incorporated in advanced vehicle control systems. In this talk I present some novel decomposition tools that allow us to untangle the infinite-dimensional dynamics of heterogeneous vehicular networks with V2V communication. These methods help us to understand the spatio-temporal complexity of large-scale multi-vehicle systems and redesign their dynamics by exploiting connectivity. Some similarities with flow oscillations on road networks equipped with V2I devices are also pointed out.

The seminar will take place at the usual time today, October 25, 2013 from 4:00-5:00 PM in 534 Davis. Cookie Hour will kick off in the library at 3:30 PM. 

Friday Seminar: Autonomous Vehicles

IMG_4915

Tomorrow's TRANSOC Friday Seminar features University of Texas Professor Kara Kockelman presenting, "Autonomous Vehicles: Anticipating Impacts in a World of Increasingly Shared Mobility." 

Autonomous vehicles (AVs) represent a potentially disruptive and beneficial change to the way in which we travel. This new technology will impact roadway safety and congestion, air quality and traveler choices. We estimate the private benefits of each AV (to individual owners) to be on the order of $2,000 per year in the near term, rising to $3,000 eventually, thanks to crash savings, travel time reductions, fuel savings, and parking benefits. When crash savings for others are included, net social benefits are estimated at over $6,000 per AV.

Nevertheless, many barriers to AV implementation and mass-market penetration exist. Initial costs will be too high for most buyers, and U.S. licensing and testing standards are being developed at the state level, rather than under a national framework, which may lead to inconsistencies across states. Liability details remain undefined, security concerns linger, and, absent new privacy standards, a default lack of privacy for personal travel may become the norm. Finally, many impacts, interactions with other components of the transportation system, and implementation details remain uncertain for this new and exciting technology.

This seminar also examines the design and results of an agent-based model for Shared Autonomous Vehicle (SAV) operations, including environmental impacts of a fleet of shared and self-driving vehicles. The model generates trips throughout a grid-based urban area, to mimic realistic travel patterns and departure times. An initial model run estimates the SAV fleet size required to reasonably service all trips, over a 24-hour period. Next, the model is run over 100 days, with driverless vehicles ferrying travelers from one destination to the next. During each 5-minute interval, some unused SAVs relocate to shorten wait times for next-period travelers.

Model applications vary trip generation rates, trip distribution patterns, network congestion levels, service area size, vehicle relocation strategies, and fleet size. Preliminary results indicate that each SAV can replace around eleven conventional vehicles, while adding up to 10% more travel distance than conventional trip-making, resulting in overall beneficial emissions impacts, once fleet-efficiency changes and embodied (versus in-use) emissions are assessed.

The seminar will be from 4-5 PM  in 534 Davis. Cookie Hour will be at 3:30 in the library. 

Urban Gridlock

Chicago Gridlock

Gridlock is a fact of life in urban areas. Why is that? A new study explores the characteristics of urban gridlock, to better understand the condition and ways to ease congestion. From Transportation Research Part C: Emerging Techonologies, "Urban network gridlock: Theory, characteristics, and dynamics" by Hani S. Mahmassani, Meead Saberi, and Ali Zockaie tackles the issue. 

This study explores the limiting properties of network-wide traffic flow relations under heavily congested conditions in a large-scale complex urban street network; these limiting conditions are emulated in the context of dynamic traffic assignment (DTA) experiments on an actual large network. The primary objectives are to characterize gridlock and understand its dynamics. This study addresses a gap in the literature with regard to the existence of exit flow and recovery period. The one-dimensional theoretical Network Fundamental Diagram (NFD) only represents steady-state behavior and holds only when the inputs change slowly in time and traffic is distributed homogenously in space. Also, it does not describe the hysteretic behavior of the network traffic when a gridlock forms or when network recovers. Thus, a model is proposed to reproduce hysteresis and gridlock when homogeneity and steady-state conditions do not hold. It is conjectured that the network average flow can be approximated as a non-linear function of network average density and variation in link densities. The proposed model is calibrated for the Chicago Central Business District (CBD) network. We also show that complex urban networks with multiple route choices, similar to the idealized network tested previously in the literature, tend to jam at a range of densities that are smaller than the theoretical average network jam density. Also it is demonstrated that networks tend to gridlock in many different ways with different configurations. This study examines how mobility of urban street networks could be improved by managing vehicle accumulation and redistributing network traffic via strategies such as demand management and disseminating real-time traveler information (adaptive driving). This study thus defines and explores some key characteristics and dynamics of urban street network gridlocks including gridlock formation, propagation, recovery, size, etc.

The full paper can be found here.

Bay Area Traffic Decoded: What cell phone and GPS data reveals about traffic patterns

580 and I80 Traffic Jam

A new article from researchers at MIT and UC Berkeley uses data from cell phones and GPS to track traffic patterns. "Understanding Road Usage Patterns in Urban Areas" from December's Scientific Reports assess how drivers from certain areas effect the whole network. 

 We find that the major usage of each road segment can be traced to its own - surprisingly few - driver sources. Based on this finding we propose a network of road usage by defining a bipartite network framework, demonstrating that in contrast to traditional approaches, which define road importance solely by topological measures, the role of a road segment depends on both: its betweeness and its degree in the road usage network. Moreover, our ability to pinpoint the few driver sources contributing to the major traffic flow allows us to create a strategy that achieves a significant reduction of the travel time across the entire road system, compared to a benchmark approach.

Drivers from Sanjose, Hayward, Dublin, San Rafel and San Ramon often find themselves stuck in the worst traffic. Could more metering be the answer?

California OKs Driverless Cars

HDR Remap

This week Gov. Jerry Brown signed SB1298, legalizing driverless cars in California. Some people are concerned about the safety risks of these robot cars. At the signing, Google's Sergey Brin said, "You can count on one hand the number of years it will take before ordinary people can experience this." Google's been testing their driverless cars for miles and miles. (For more background, see Sebastian Thrun's TED Talk.)

Autonomous vehicle research has been progressing for decades. It was part of USDOT's IVHS (Intelligent Vehicle and Highway Systems), what we now know as Intelligent Transportation Systems. For the history of research in the field, just look at TRID and how certain concepts have evolved. Autonomous land vehicles (Qbddkmb) shows the clear relationship to some military efforts. The same can be said for Autonomous vehicle guidance (Dcmvgyh). Connected vehicles are a bit trickier since you have to look at Vehicle to roadside communications (Dsbnu) and Vehicle to vehicle communications (Dsbnw), and Vehicular ad hoc networks as well.

So we might not have Herbie the Love Bug quite yet, but researchers are bringing us quite close to it.

 

New Global BRT Database

Estação e vermelhão

This week BRTdata.org was launched by the Bus Rapid Transit Center of Excellence and EMBARQ. The site acts a clearinghouse for data from BRT systems all over the world. You can see performance indicators by country or city, such as passengers per day, number of corridors, and legth. Check it out and let them know what you think.

How much longer do we have to wait for cars that drive themselves?

Google Self-Driving Car

Today Greater Greater Washington blogged about the prospect of self-driving cars.

Whether we are prepared for it or not, the next revolution in transportation will be here soon, and it won't be streetcars, monorails, segways, or electric vehicles. It will be self-driving cars, and the adoption of this technology will change everything we accept as a given in the field of transportation planning.

They also link to a Washingtonian interview with Michael Pack, director of the CATT Laboratory at the University of Maryland, and noted transportation technologist. He sees autonomous vehicles as a potential solution for congestion, "Completely automated cars that take the driver out of the equation, communicate with one another, and can travel at high speeds within six inches of one another."

Will Hansfield on Greater Greater Washington projects that we might see self driving cars commercialy viable in the US in the next 7-12 years. Given the clip of research, it might not be far off. Looking at "intelligent vehicles" research in TRID, automation is trend that has been becoming more common over the years. From cyber cars to intercontinental van journeys, integrated systems for autonomous vehicles are coming.

The PATH program from ITS Berkeley has been looking at autonomous vehicles for quite a while now. Though the most famous driverless cars might be the new fleet from Google. Sorry KITT.

Syndicate content