BeiDou. Construction of the second phase of BeiDou has been completed; further launches for the third phase – constellation completion – are on hold until tests of the existing 14-satellite constellation are complete, according to Xiancheng Ding, Senior Advisor, China Satellite Navigation Office. As of December 27, 2012, BeiDou achieved full operational capability for most of the Asia-Pacific region. The full constellation is now expected to be completed by 2020.

Other accomplishments include releasing the BeiDou Interface Control Document and manufacture of BeiDou chips for end-user applications. By the end of June, some manufacturers will release BeiDou chips in China, Ding said.

Also in December, BeiDou introduced a new logo (at right).

Yuanxi Yang (China National Administration of GNSS and Applications) presented statistics showing that BeiDou+GPS provides greater accuracy than GPS alone. For instance, the RMS of BeiDou+GPS kinematic positioning by using differential carrier phase is about 20 percent better than that of GPS alone, Yang said.

By itself, existing BeiDou constellation system accuracy is better than 10 meters, timing better than 20 nanoseconds, and velocity accuracy is better than 0.2 meters/second.

In all, BeiDou is composed of 14 satellites: five GEO, five IGSO, and four MEO. The full constellation (by 2020)  will consist of 35 satellites: 5 GEO and 30 non-GEO (a mixture of MEO and IGSO satellites).

GPS. Keynote speaker David A. Turner (U.S. Department of State) shared his time with surprise GLONASS speaker Sergey Revnivykh (International Committee on GNSS, ICG). In his GNSS Policy and Program Update, Turner provided the dates by which three new civil signals will be on 24 GPS satellites.

• The L2C signal is a developmental signal broadcasting from 10 GPS Satellites. It began launching in 2005 with GPS Block IIR(M) satellites, and is expected to be available on 24 satellites around 2018.
• The L5 signal is a developmental signal broadcasting from three GPS satellites. It began launching in 2010 with Block IIF satellites, and is expected to be available on 24 GPS satellites around 2021.
• The L1C signal begins launching in 2015 with GPS III; available on 24 GPS satellites around 2026.

“We have an increasing number of signals, increasing capability, and increasing level of service as we continue to evolve the constellation,” Turner said.

GLONASS. The next GLONASS satellite will be launched this Friday, April 26, Revnivykh said. This will be a GLONASS-M satellite, number 47. The first launch of a new generation GLONASS K satellite is scheduled for 2015.

Revnivykh stressed GLONASS’ role as a global utility. “We consider international cooperation is essential for all GNSS, and we consider GLONASS an essential part of the international multi-GNSS system,” he said. He stressed the importance of compatibility and interoperability as key to this policy.

In 2012, GLONASS performed with an average accuracy better than formally required, he said. GLONASS is in worldwide use, and positioning has improved by a factor of 10, from 35 meters to about 3 meters since the first satellites were launched. Using both GPS + GLONASS provides 1.5 times better high-precision measurements, Revnivykh said.

The new GLONASS program for 2020 for GLONASS sustainment, development, and use includes GLONASS M, K1, and K2 satellites; the positioning accuracy objective is to go from the current 2.8 meters to 0.6 meters.

Aviation. Chris Hegarty (MITRE) presented an FAA Navigation Programs Overview on behalf of the scheduled speaker Deborah Lawrence (FAA) who was unable to attend. He noted that United Airlines has begun GBAS operations in Houston.

In answer to a funding question, he said, “The sequestration is not expected to have a positive effect on schedule, but the presented timeline for APNT is the FAA’s current best estimate. Congress has some tough decisions before them, and I wouldn’t want to speculate on potential schedule impacts. In the words of Yogi Berra, predicting is hard, especially when it involves the future.”

Korean SBAS. Changdon Kee (Seoul National University) shared plans for a Korean SBAS. In South Korea, LPV availability is 49.4% compared to 90.6% in Japan. “Korea needs its own system,” Kee said.

Phase 3 of the SBAS development could start by the end of September, depending on funding. It will include open service multifunctional GEO satellites interoperable with other SBASs. A pseudolite demonstration system will be completed in 2014, clearing the way for the beginning of Phase 3.

In all, the system will include five reference stations, two master stations, two ground uplink stations, and two GEO satellites (the main GEO by 2018 and a backup by 2020).

The Korean SBAS open service system will provide GPS L1 augmentation, begin operation in 2020, and support aviation, land and maritime users. A test operation system will provide GPS L1 and L5 augmentation. The system is expected to be fully operational by 2021, with service available throughout Asia.

Japan’s QZSS. Hiroyuki Noda (Office of National Space Policy, Japan) said three more satellites for this augmentation system will be launched by the end of the decade, with the service beginning in 2018. In September 2012, the Japan cabinet made the commitment to accelerate development of the system. The first satellite, launched in 2010 (QZS-1, aka Michibiki) is performing as expected.

QZSS is expected to improve positioning availability from 90% to 99.8% in Japan. QZSS will not only improve positioning in the Asia-Pacific region, but is expected to improve the capacity to respond to natural disasters, Noda said.

Surveyors install and configure a base and rover for a 13,000-hectare survey of the Plains Kogoni in Mali.

In the heart of landlocked Mali, between the Atlantic Ocean 800 miles to the south and the Sahara desert to the north, lays the extraordinary Inner Niger River Delta, also known as the Macina, a 1.8 million hectare oasis of lakes and floodplains with a vast potential for hydro agriculture.

CIRA, a major West African consulting engineering firm, working on behalf of the Office du Niger, a quasi-governmental Mali company charged with managing more than100,000 hectares of irrigated delta land, has completed surveying an additional 25,000 hectares for hydro-agriculture development.

Created in 1991, CIRA is an engineering and applied research consulting firm working in transportation, hydraulics, civil engineering and the environment. Based in Bamako, Mali, the firm works in more than 15 African countries, primarily in West Africa, Central Africa and East Africa.

In the course of two months during the dry season, two CIRA survey teams, each equipped with three Spectra Precision ProMark 500s, a base station, and two rovers connected via UHF, completed the entire 25,000 hectare survey collecting four points in x, y, and z per hectare to produce a digital model. The model enabled the production of rough pre-study with all plans and a detailed pre-project CAD drawings for drainage, irrigation canals, and related infrastructures.

A very short eight-month contractual time set to complete the different studies meant that the land survey study would have to be completed as quickly as possible. The first thought was to use aerial photography combined with LIDAR, but setting this up would have taken too long, according to a CIRA spokesperson. Instead, CIRCA chose to employ differential GNSS, using base and rovers working in real-time kinematic. CIRA’s experience suggested the firm would achieve reliable results much quicker using only optical total stations. CIRA elected to use Ashtech ProMark 500 GNSS receivers for the project. From experience, they knew the models were easy to set up and use, lightweight, offered long battery life in the field, and field to office data transfer would be easy. Their expectations were met, and the job was completed within two months and on time.

The ProMark 500 RTK survey system provides short time to fix, long-range RTK and solution reliability. Its BLADE technology provides multi-constellation signal processing with the use of SBAS and GLONASS ranging signals to strenghten the GPS solution.

Trimble acquired Ashtech in 2011, making it part of Spectra Precision.

Setting up bitter points for calibration of satellite images on the corridor Sarh – Abeche in Chad (800km).

Reference station during the survey topo Richard Toll road – N Dioum (120 miles) in Senegal.

A reference station during the survey topo Zégoua Sikasso road (95 km) in Mali.

UC Berkeley researchers have developed a method to provide real-time, high-resolution data in hard-to-map waterways, using GPS. Tossing a robot is Andrew Tinka, with Kevin Weekly. (Photo courtesy of Jérôme Thai.)

By Tracy Cozzens

A fleet of 100 robots equipped with GPS and sensors were released May 9 into California rivers to measure water flow, salinty levels, and pollution. The Floating Sensor Network is a project by the University of California, Berkeley, to improve the way water quality and flows are monitored.

About two-thirds of California’s fresh water is in the Sacramento-San Joaquin river system where the test took place. This water supplies about two-thirds of the state’s population with drinking water and irrigation. The initiative is led by associate professor Alexandre Bayen at the Center for Informatin Technology Research in the Interest of Society (CITRIS).

The robots each have a sensor to test salinity and a GPS unit from a smartphone. Some have propellers so they can maneuver around obstacles and reach specific destinations. The robots also sent Tweets to @fsnandroid61.

The robots drifted through the area of the river being measured, then were retrieved by boat. “One advantage of our real-time communication system is that we can see where all our sensors are on a map, which makes it very easy to chase them down and retrieve them,” said graduate student researcher Andrew Tinka.

With the first test completed, the team’s efforts over the summer have two priorities, Tinka explained. “First, we’re using the flow data that we gathered on May 9 to understand how this ‘mobile’ data can be best used for river hydrodynamics studies. We’re learning how to turn the individual traces of water that each sensor gives us into a big-picture view of the entire river region, sort of how like meteorologists take the data from a few weather stations and turn it into an overall view of what the weather is doing over a large area. Second, we’re working with other hydrodynamics research groups to expand the use of this kind of mobile sensor. We’re loaning our equipment to other groups, doing pilot projects with others, and basically trying to get these sensors into researchers’ toolboxes throughout the water community.”

There are two types of devices in the fleet, active and passive. The active sensors have a twin-propeller drive system that lets them move through the water to avoid obstacles or stay in the correct region of the river. “We developed the internal electronics for this device ourselves,” Tinka said. “We integrated a Magellan AC12 GPS receiver along with a Gumstix embedded computer and a Motorola GSM module. Our passive sensors don’t have a propulsion system; they do exactly what the water does. We developed this system with a focus on cost and ease of assembly.” The team used a waterproof consumer smartphone, the Motorola Defy, for the GPS positioning, computation, and comunications.

So far, the test has proven the usefulness of such a network. The devices were developed to be easily deployable, especially where a lot of flexibility is needed, such as in disaster response. “The ability to quickly and easily put these sensors into new inland environments, by just about any method (throw them from a boat, drop them from a helicopter, toss them from a dock or a bridge) makes them a really useful new tool,” Tinka said.

 
A hundred robots, 40 with propellers, were released into the Sacramento River near Walnut Grove (photos courtesy of Jonathan Beard).

Using a large network of GPS stations, a team of researchers has found that the Rio Valley Rift in the Southwest United States — previously suspected to be dead — is slowly expanding, at a rate of about 0.1 millimeter per year.

The Rio Grande Rift extends from Colorado’s central Rocky Mountains to Mexico.

The study was conducted by scientists at the Cooperative Institute for Research in the Environmental Sciences (CIRES) at the University of Colorado at Boulder, in collaboration with the University of New Mexico, New Mexico Tech, Utah State University, and UNAVCO.

“We don’t expect to see a lot of earthquakes, or big ones, but we will have some earthquakes,” said study author Anne Sheehan, CIRES Fellow and associate director of CIRES Solid Earth Sciences Division. “We use continuous measurements of GPS sites from across the Rio Grande Rift, Great Plains, and Colorado Plateau to estimate present-day surface velocities and strain rates,” Sheehan said.

Using GPS instruments at 25 sites in Colorado and New Mexico, the team tracked the rift’s miniscule movements from 2006 to 2011. The team found an average strain rate of 1.2 nanostrain each year across the experimental area. A nanostrain is a change in length of one part per billion, thus 1.2 nanostrain per year is equivalent to 1.2 millimeter per year extension over a 1000-kilometer length.“If you picked two points in New Mexico, and one of them lies 100 kilometers to the west of the other, then they would be moving apart at a rate of 0.1 millimeter per year,” explained researcher Henry Berglund.

Researchers used data from 25 continuous GPS stations installed as part of the EarthScope Rio Grande Rift GPS experiment, supplemented by data from other GPS monuments in the southwestern U.S., resulting in a data set of daily position estimates of 284 GPS monuments for the years 2006 through 2010.

“It is lower than we thought but it does exist,” Sheehan said. “Some people thought it was zero but we are seeing things are extending slowly.”

The slow rates of motion made previous attempts to determine tectonic activity difficult. Previously, geologists had estimated the rift had spread apart by up to 5 millimeters each year but the errors introduced by the measuring instrumentations were significant. “The GPS has reduced the uncertainty dramatically,” Sheehan said. “This is the most comprehensive and accurate set of geodetic measurements in this area to date.”

The extensional deformation is not concentrated in a narrow zone centered on the Rio Grande Rift. Instead, it is distributed broadly from the western edge of the Colorado Plateau into the western Great Plains — a span of more than 370 miles. “This unexpected pattern of broadly distributed deformation at the surface has important implications for our understanding of how low strain-rate deformation within continental interiors is accommodated,” Sheehan said. “Questions we wanted to answer are: how is the Rio Grande Rift deforming? Is it alive or dead? Is it opening or not?”

Along the rift, spreading motion in the crust has caused magma to rise to the surface, creating long basins susceptible to earthquakes. “The rift is still active,” Sheehan said.

The team plans to continue monitoring the Rio Grande Rift, and may attempt to determine vertical as well as horizontal activity to determine whether the Rocky Mountains are still uplifting.

University of Colorado (Boulder) student Henry Berglund services GPS site RG20 west of Silverton, Colorado.

The study’s findings shed light on how continents deform away from plate boundaries, Sheehan said. At plate boundaries scientists can clearly see what is going on. “Things move past each other and crash into each other. At active plate boundaries, the rates of motion detected by GPS can be centimeters per year. Compare that with the fraction of a millimeter per year that we have measured for the Rio Grande Rift.”

“Present day measurements of deformation within continental interiors have been difficult to capture due to the typically slow rates of deformation within them,” Berglund said. “Now, with the recent advances in space geodesy, we are finding some very surprising results in these previously unresolved areas.”

The National Science Foundation funded the study. EarthScope and UNAVCO provided instruments, equipment, and engineering services. Results of the study were published in the January 2012 issue of Geology magazine.

GPS monuments in the vicinity of the Rio Grande Rift and southern Rocky Mountains. The study included construction of 25 GPS monuments (blue circles) in Colorado and New Mexico in 2006 and 2007. Regional EarthScope Plate Boundary Observatory and Continuously Operating Reference Station monuments are shown by gray triangles.

By Tracy Cozzens

Beneath the surface of a tropical paradise in the city of Townsville on Australia’s Sunshine Coast lies a hidden maze of tunnels and underground bunkers, once said to be used by General Douglas MacArthur. Learning the secrets of this labyrinth that was a major World War II staging point for battles in the Southwest Pacific is the passion of Kevin Parkes of Geo Positioning Services, Townsville.

Parkes’ main tool is historic aerial photography, coupled with hours of research in the National Australian Archives and the National Library of Australia. To that he adds geophysical surveys of the infrastructure. Parkes is undertaking the geophysical surveying and mapping using an Ashtech ProMark 100 GNSS receiver and a Willy Bayot PPM Mk 3 magnetometer. He used the magnetometer and GPS receiver in parallel, later processing both data sets.

After the attack on Pearl Harbor and the Japanese advance through Asia, Townsville’s population bloomed from 30,000 to 120,000 by mid-1943. The rapid military influx stretched resources to the breaking point.

Landing Ship Tanks discharge on the strand in Townsville.

The U.S. Army 5th Air Force established the largest aircraft repair and maintenance facility ever built in the southern hemisphere at Townsville, and the site became the technical hub of U.S. military aviation. Air Force Service Command Depot #2 at Townsville was capable of overhauling 300 aircraft engines per month and performed aircraft assemblies, modifications, overhauls, and maintenance. Major resources and facilities serviced the Royal Australian Air Force, Australian and U.S. Armies, Royal Netherlands Air Force, Royal Air Force, Canadian forces, Royal Navy, and other allied forces.

“A visitor to Townsville today would be forgiven in asking where the artifacts of this massive military facility are today,” Parkes said. “There is very little remaining in any built structures that give any idea of what happened in this city 70 years ago.”

Parkes realized that underground cave shelters were most likely used for warehousing and storage, to keep stores out of the weather and protected from enemy action.

He describes one area he investigated, a park in Townsville used as an officer’s accommodation camp. Preliminary magnetic anomaly surveys indicated linear anomalies were beneath the park surface. A high-resolution survey gave samples of about 1.5-meter resolution.

“The difficulty was reducing all noise levels down to a minimum, including the X/Y positioning, so the GPS requirements came down to survey quality,” Parkes said. “It is absolutely critical that the GNSS receiver and magnetometer keep in synchronization during data collecting runs including under the frequently encountered tree canopies.”

To improve accuracy, Parkes avoids using real-time kinematic survey equipment. “That would involve having another electronic device operating and emitting more noise in the signal spectrum,” he said. The need to position the GPS antenna in close proximity to the magnetometer sensor was a major issue with all on-pole RTK systems.

A U.S. Army air raid shelter under the officer’s accommodation camp, mapped with GPS and magnetometer data and using Surfer 3D surface mapping software.

With an Ashtech Promark 3, post-processed results were better than 100-millimeter X/Y coordinates. “The unit is lightweight and self-contained,” Parkes said. “The noise from the Ashtech survey-grade external antenna’s effect on the magnetometer data was insignificant.”

Still, this park had a grove of trees that defied every attempt to maintain GPS reception and consequently synchronize the magnetometer. Along came the Ashtech ProMark 100, a lightweight and self-contained receiver with external geodetic antenna with GPS and GLONASS. “My first attempt at surveying under the trees was spectacular to say the least,” Parkes said. “Synchronization with the magnetometer data was near perfect.”

The dual-constellation reception of the ProMark 100 became essential to the success of Parkes’ work. After more than a hundred data-collection passes with the magnetometer and ProMark 100 through the groves of trees, at no time did the Position Dilution of Precision (PDOP) rise to more than three, and at all times more than eight satellites were available. The ProMark 100 data is post-processed to improve accuracy. Parkes noted that ironically many of the most interesting finds have been collected under heavy tree canopy. Without the quality of the geographic positions enabled by the ProMark100 under tree canopy, Parkes said that much of his work would have been impossible to achieve.

Parkes’ surveying equipment includes a magnetometer and a ProMark 100 GNSS receiver.

In fact, when Parkes first began his mapping project in 2005, he used a single-constellation GPS system and post processed the results against the local International GNSS Service (IGS) reference station. The GPS-only system worked very well until a grove of trees would interfere with the sky. Now with the ProMark 100 GNSS receiver, Parkes surveys using GPS L1 and GLONASS in continuous kinematic mode at a one-second collection rate. He then post processes the data against another ProMark 100 used as a local reference station.

To date, Parkes has mapped an underground railway, artillery observation posts, several shelters, fuel terminals and other yet-to-be-identified pieces of the vast infrastructure.

During his Research, Parkes mapped a major magnetic anomaly in Cleveland Bay. In 1770 Captain James Cook in the HMS Endeavour mapped the east Australian coast. Venturing into Cleveland bay, Cook noticed his compass behaving erratically, and named one island Magnetic Island. Today, a 3D surface model reveals a large magnetic anomaly heading across Cleveland Bay and straight towards Magnetic Island, 7 kilometers from Townsville. Experts who have examined the data believe that it is a naturally occurring magnetic anomaly about 800 meters wide. “It would appear that Captain James Cook was indeed a very capable navigator and cartographer,” Parkes said.

Nav On Time, a French Company located in Toulouse, has successfully completed a trial campaign of its Mow-By-Sat precision guidance on a commercial lawnmower. In August, the prototype of a GPS-guided robot lawnmower was installed on a golf driving range near Toulouse and tested in real conditions of use, day and night, maintaining a 25,000 square meter lawn since then. In a previous campaign, the mower covered more than 2.2 million yards — equal to1,250 miles or 2,000 kilometers — in 2,100 hours. (See videos of the mower in action at www.youtube.com/DSnavontime.)

With such a success under its belt, Nav On Time is negotiating with different lawnmower manufacturers to bring a product to market. The autonomous lawnmowers already on the market, including machines commercialized by research partner BelRobotics, use underground wired perimeters for delimiting the lawn by an electromagnetic signal, the strength of which is measured by a mower-embedded sensor to determine its distance to the lawn’s limit. But that wire, and its required installation, are technical barriers for a lot of potential customers. Nav On Time is one of the companies developing solutions to get rid of the perimetric wire yet still be able to guide the mower autonomously with accuracy and efficiency.

The Mow-By-Sat prototype was developed under an R&D project with Belrobotics, maker of the Bigmow commercial mower.

Between January 2009 and June 2010, Nav On Time coordinated the Mow-by-Sat project, a research and development effort that received funding from the European Union’s Seventh Framework Programme (FP7/2007–2013). Partners included Belrobotics of Belgium, a large lawn-maintenance robot manufacturer, and the University of Catania in Sicily, Italy, through its robotics research department.

The Mow-by-Sat project (www.mow-by-sat.eu) was also undertaken to support development of a GNSS-based navigation and guidance system integrated into an autonomous lawnmower, paving the way for industrialization and commercialization of GNSS applications for a domestic service robot operating outdoors. Beyond this concrete application, the project aimed to increase the adoption of GNSS technologies in robotics applications, studying the benefits of European GNSS (especially EGNOS and Galileo).

Mow-By-Sat uses a virtual fence to replace the wired boundary traditionally used in robot lawnmowers, which provides better flexibility for defining and modifying a mowing area. Mow-By-Sat enhances the machine’s efficiency by a factor of three, as full steering substitutes for the random operation mode, the company said.

Built around a European GNSS L1 automotive receiver, the u-blox T, Mow-By-Sat uses L1 fixed / floating real-time kinematic (RTK) techniques. A tight coupling between the RTK positioning firmware and the guidance application software aids the mower’s precision. Nav On Time compared it to the challenges of aviation, where the required navigation performance depends on the flight phase.

In its patented architecture, the module embedded in the rover is dumb, and the ground-based station acts as a remote control, ensures traffic management between several machines, and serves as a gateway for remote services such as installation, supervision, and surveillance, all accessible from the Internet. Nav On Time developed both the positioning firmware and guidance application software.

According to Nav On Time CEO Michèle Poncelet, Mow-By-Sat offers significant competitive advantages to the machine manufacturer compared to expensive RTK solutions now on the market. She cited:

• easy customization because of its open architecture,
• an affordable solution for small and inexpensive mobile machines,
• a technology enabler for replacing human-controlled and energy-consuming machines with smaller and cheaper machines that have a smaller carbon footprint.

With six Engineers, Nav On Time, founded in 2007, is offering a product line dedicated to precision control solutions for small and inexpensive mobile machines, under a business-to-business model through industrial partnerships. According to Poncelet, its market stretches from human controlled machines (precision agriculture or crane collision avoidance) as driver’s assistance, to unmanned machines (autonomous lawnmowers, other unmanned ground vehicles, intelligent vehicles, and more generally service robots) with full steering.

Other applications envisioned by Nav On Time include a golfball retrieval robot for driving ranges, a beach cleaner robot, and a surveillance robot — any application that requires passing through a pre-determined area in a methodical and systematic way.

In a rigorous test, the Mow-By Sat kept a golf driving range tidy, never complained,
and never grew bored with mowing, unlike its human counterparts.

Breaking Ground

It would seem mowing lawns isn’t a beloved pastime, as autonomous lawn mowers have been the subject of numerous research projects. For the past eight years, the Institute of Navigation has sponsored a Robotic Lawnmower Competition as a way to encourage college students to develop autonomous steering techniques. During the second ION Autonomous Lawnmower competition, Frank Van Graas, who accompanied the winning Ohio University team, told GPS World, “The centimeter-level positioning accuracy required for lawnmowers in the contest is actually more difficult than automatically landing an airplane.”

One research project, carried out by Navcom Technology in 2005, resulted in an autonomous mower taking on the precise mowing techniques of baseball stadiums, with its checkered patterns. The Navcom project, documented by Michael Zeitzew in his paper “Autonomous Utility Mower,” used a series of beacons to augment GPS. Two off-the-shelf John Deere utility mowers were modified for X-by-wire control, and fixed navigation beacons were mounted around the stadium. Next, the field boundaries were surveyed and input into a map file, used to create the mower’s mission plan.

“The use of GPS requires good sky visibility,” explained Zeitzew. “In this application, due to the stringent navigation accuracy requirements, an RTK-GPS solution is required, which requires the use of a base station. Because many of the baseball stadiums have high walls and other obstructions around the field, RTK-GPS is inadequate, even with augmentation by (affordable) inertial sensors or odometry sensors. This necessitated the use of alternative technology.”

Navcom fielded two mower systems into professional baseball stadiums, one major league and one minor league. Both systems were used over the course of several weeks during the spring 2005 baseball season, and received positive reviews from the professional groundskeepers, who quickly grew comfortable using the machines. The project proved not only that autonomous mowers are possible even for large-scale sites such as a stadium, but that there is indeed a market for them.

GreenGPS is being developed by University of Illinois researchers.

Most GPS devices in cars today give the driver two choices: shortest route or fastest route. GreenGPS provides a third option: most fuel-efficient route.

With gas prices skyrocketing, many drivers would be happy to spend a few more minutes on the road, or take a different route, if it meant burning less gas.

The answer could be the GreenGPS navigation service, now being developed by researchers at the University of Illinois at Urbana Champaign (UIUC), which finds the most fuel-efficient route for your vehicle.

“The most fuel-efficient route may be different from the shortest route because the latter may pass through downtown stop-and-go traffic,” explained Tarek Abdelzaher, project lead and computer science professor. “It may also be different from the fastest route because vehicles are not as fuel efficient at higher speeds.”

All cars manufactured in the U.S. since 1996 come with a standard interface to their internal gauges and engine measurements called the On-Board Diagnostics Interface, or OBD-II. GreenGPS runs on the driver’s GPS-enabled cell phone and uses an off-the-shelf wireless adaptor plugged into the vehicle’s OBD-II port to receive engine readings via Bluetooth.

The cell phone collects the readings and connects to a server that models the engine’s fuel efficiency and customizes navigation advice to the particular vehicle, Abdelzaher explained.

The best route computed by GreenGPS to the same destination may differ from one vehicle to another. “For example, my vehicle uses about 20-25 percent more gas in stop-and-go traffic compared to free-flowing traffic, whereas my wife’s car uses closer to 40 percent more,” Abdelzaher said. “GreenGPS may recommend to her a path that is longer but has no traffic, whereas it might recommend to me a path that incurs some traffic but is shorter.”

To users, GreenGPS looks like a regular navigation service. The driver specifies a destination, then ask the service to find a route. “It runs on your cell phone, except that in addition to the fastest and shortest route options, it offers the ‘least-fuel route’ option,” Abdelzaher said. “If the driver chooses that option, they receive the GreenGPS-recommended fuel-efficient route.”

The program works best with a small hardware addition to collect readings specific to the vehicle. “In order for the advice to be customized to the performance of your specific vehicle, the driver should invest in buying the OBD-II adaptor. It costs about $50 and is a one-time investment,” Abdelzaher said.

“If the driver does not wish to buy the adaptor, they can still use GreenGPS and supply the make, model, and year of their vehicle. In this case, GreenGPS will use data from other vehicles of the same make, model, and year, or vehicles as close to them as possible to compute the navigation advice,” Abdelzaher said. This social networking component is also being developed as part of the project.

The system pulls the GPS data from the driver’s cellphone. “If you use a GPS phone (and most smartphones have GPS), the system simply finds out your current location from your phone. Otherwise, you would need to supply both source and destination addresses (like you would when you get directions from Google Maps) and the system will show you the route on a map.”

Gas-Saving Pilots. In the first stage of testing, the team solicited volunteers to drive in the area of their university, in Urbana-Champaign, a city of 170,000. In all, 1,000 miles were driven by 16 different cars. Results demonstrated that following the fuel-efficient route saves on average 6 percent over the shortest route, and 13 percent over the fastest. “This was done on flat terrain and in the absence of significant congestion,” Abdelzaher said. “We expect that testing in higher traffic and richer topology will increase the variability in fuel consumption among different routes, resulting in even more potential savings when following the most fuel-efficient route. Verifying this conjecture is currently a topic of investigation for our ongoing research project.”

Abdelzaher said his team has just started the second stage. “In the second stage of testing, we will deploy GreenGPS on the UIUC Facilities and Services fleet (about 100 cars) and monitor performance over a longer period of time. Preparations for this deployment are currently under way. We also expect to offer GreenGPS publicly to any other volunteers who wish to help with testing.”

Impressed by early findings of gas savings, the second phase is being funded in part by a $300,000 grant from the National Science Foundation. Abdelzaher and Robin Kravets, another UIUC computer science faculty member, were awarded the grant this spring. It will help deploy GreenGPS among the campus fleet cars, and track and analyze the results.

“By sharing data on speed, fuel efficiency, and location of vehicles, better real-time navigation services can be developed that guide drivers to routes that are maximally fuel-efficient for their cars, hence reducing transportation carbon footprint,” the grant reads. “This project helps usher in a new era of sensing applications with more integration of humans, networks, and the physical world, which may have a significant impact on the economy, energy, and the environment by reducing transportation energy cost and carbon footprint.”

Other grant providers are the Office of Naval Research, which is funding research on the technology’s networking component, and IBM through its Smarter Planet initiative. As a part of this project, 200 or more cars in the Urbana-Champaign area of various makes and models will be fitted with GreenGPS. Through a social network of drivers, data and routes collected can be shared and used by those who don’t have the OBM-II adaptor installed.

Fuel consumptions for the various roundtrips between different landmarks.

The landmarks and corresponding shortest and fastest routes.

Figure 1. Worldwide nuclear testing 1945–2009 (CTBTO website).

Can GPS be used to detect underground nuclear explosions?

A research team is developing a software program that uses GPS to analyze the ionospheric effect of nuclear explosions. Results would show when and where a country has conducted a secret underground nuclear test. Team members are Jihye Park, Ralph. R. B. von Frese, and Dorota A. Grejner-Brzezinska from The Ohio State University and Jade Yu Morton from Miami University.

The Comprehensive Nuclear-Test-Ban Treaty was adopted by the United Nations General Assembly in 1996, but not all nuclear countries have ratified it, including the United States, China, Egypt, Indonesia, Iran, and Israel. Also, India, North Korea, and Pakistan have not signed the treaty.

Park, a doctoral student in geodetic science at Ohio State, created the computer program to detect changes in the ionosphere from nuclear weapons testing.

A previous study showed that the ionosphere was disturbed by underground nuclear testing conducted by Russia in 1990. GPS is capable of precisely measuring the total electron content (TEC) of the ionosphere along the path between satellite and receiver at a GPS station, so Park and her team decided to begin researching the use of GPS in detecting nuclear explosions.

“Many studies have been done to monitor and model the atmosphere using GPS technology,” Park said. “Research has proven that GPS can detect natural disasters such as earthquakes or tsunamis. This study broadens those areas of study with its capability to detect underground explosions.”

Detonation of a nuclear weapon results in a shockwave that travels through the atmosphere, changing the density of charged particles in the ionosphere. “The explosions can’t hide from the ionosphere,” said von Frese, geophysicist and project leader. “Our technology would be another nail in the structure to detect explosions.”

“One of the arguments is ‘Well, how do you prove that a clandestine explosion occurred?’” said Grejner-Brzezinska, Park’s adviser and GPS World’s Tech Talk blog editor. “Now we can say, ‘Here, we have the data from GPS to show when and where.’”

According to the Comprehensive Nuclear-Test-Ban Treaty Organization (CTBTO) nuclear testing has been carried out in the past by the United States, Russia, the United Kingdom, France, China, India, Pakistan, and North Korea (see Figure 1).

Researchers, and those monitoring treaty violations, are able to target specific geographic areas that are equipped for tests, since development of a nuclear test site requires a lot of technical effort and budget. For example, the North Korean tests carried out in 2006 and 2009 were very close geographically.

“They tend to stick to the same site and reuse their facilities for nuclear testing,” von Frese said. “So a country that has previously conducted underground nuclear testing probably will reuse the site if new testing is needed.”

“They could be monitored using GPS as long as there are GPS stations nearby,” Park said.

The new GPS nuclear-detection technology was presented at the Comprehensive Nuclear-Test-Ban Treaty Organization meeting held June 8–10 in Vienna, Austria, and received press coverage that drew additional interest.

GPS Detection. The team zeroed in on a specific event to test the software, selecting a nuclear test conducted by North Korea in 2009 and using data pulled from nearby South Korean GPS stations.

Traditional detection methods for underground nuclear tests include seismic and other sensors. The CTBTO operates an international monitoring system to detect explosions with a yield of at least one kiloton. Besides seismic sensors, monitoring includes hydroacoustic sensors to monitor for shockwaves on land and in water, infrasound to detect pressure waves, and radionuclide detectors for any gas that may have been generated, though the levels aren’t always detectable.

“Even though there are four different systems available, they sometimes are unable to detect the underground nuclear explosions,” Park said. “GPS technology will make the detection validation stronger since each of them is based on a different theory. In the case of the nuclear test conducted by North Korea in 2009, only seismic and a few infrasound sensors detected the event because of their improved containment technique. Our study tracked down the 2009 event using GPS, and found it coincided with the seismic results.”

Park was able to take advantage of the well-established worldwide infrastructure already in place for GPS for her software test. The team used GPS data recorded by South Korean GPS receivers of the 2009 North Korea test. “There are a few IGS (International GNSS Service) stations in South Korea, China, and Japan. Since South Korea runs their own GPS network, I requested the data so that we could obtain data from more stations located in South Korea,” Park said.

“Since the stations we chose were permanent reference stations controlled by an international organization (IGS) and a specific country (Republic of Korea or South Korea) respectively, most of them have been running continuously except for unexpected data gaps from time to time,” Park said. Figure 2 shows the GPS stations processed for the project.

With data in hand, Park was able to test her software. The results showed definite peaks from different stations at different times after the 2009 explosion. “We realized that the time of the detected peak was dependent on the distance between underground nuclear explosion and each GPS station,” Park said. Figure 3 shows four different stations’ TIDs (traveling ionospheric disturbances) that the team initially recognized.

Figure 3. Traveling ionospheric disturbances (TIDs) detected at stations INJE (top left), DOND (top right), DAEJ (bottom left), and CHAN (bottom right). Click to enlarge.

Ruling Out Quakes. One big challenge using GPS for ionospheric monitoring is determining the origin of an event. “Since earthquakes also disturb the ionosphere, distinguishing earthquakes from underground nuclear explosions are problematic even with GPS,” Park said. “Indeed, we only focused on examining and isolating TIDs from the nuclear explosions. We are now working to analyze the TIDs from earthquakes and compare them with nuclear TIDs.”

Besides helping to distinguish between earthquakes and nuclear-test explosions, the software may eventually distinguish between nuclear plant fallout and nuclear test fallout.

With this goal in mind, the team is analyzing the ionospheric data gathered from recent nuclear plant accidents such as the one in Japan following the earthquake and tsunami in March. “Since there were data gaps and other data issues, we have as yet nothing more to report. Hopefully, we find the earthquakes’ signature soon.”