MACHINE-MADE ARMOR

One widely recognized element of the traditional Japanese armor worn by Samurai warriors is the ornate helmet—called the kabuto—adorned with family crests, sculptures, and mythical images.

Japanese manufacturer Tomco machined this
kabuto helmet through the use of CAD, CAM,
and laser scanning.

Tomco, a manufacturer in Tokyo that carries out high-speed computer numerically controlled machining, has manufactured a scale model of a kabuto. The model demonstrates the intricate details that CNC machines can execute, according to a Tomco spokesperson.

Starting with an original kabuto, Tomco engineers captured the shape with a laser scanner before manipulating the model in Catia CAD software. CNC programming was done in the WorkNC CAM and CAD package from Sescoi of Southfield, Mich.

Working from a solid block, machinists used the CAM program to machine the details on the outside of the helmet. The process was then repeated on the inside of the kabuto, according to Tomco.

SEA SEARCH HELP

British and U.S. researchers are developing a new computer model to predict how long someone will survive when lost at sea, to determine when a search-and-rescue operation should be stopped.

Researchers at the University of Portsmouth’s Sports Science and Mathematics departments in Portsmouth, England, and the U.S. Coast Guard are at work on the Search-and-Rescue Survival Model, said Professor Mike Tipton, a project researcher. He’s a human and applied physiologist at the University of Portsmouth.

“Using this new computer model will take pressure off humans making very emotional and sensitive decisions about when to end a search,” he said. “When the model predicts that a victim can no longer survive, the search-and-rescue team can stop or redeploy the search.”

The model aims to support the Coast Guard target of saving 93 percent of victims annually—from 3,000 to 6,000 people—whose lives are in danger in U.S. waters, said Chris Turner, ocean engineer and manager for the project for the U.S. Coast Guard in Groton, Conn.

The Coast Guard currently uses a software system known as the Search-and-Rescue Optimal Planning System, or SAROPS, to calculate how far a person will have drifted and how large the search area should be, Turner said.

Data such as wind speed, sea state, and water temperature is entered into the system along with information about the victim’s sex, height, weight, and clothing to determine exactly how the search should be conducted, Tipton said.

“Calculating survival time will add another layer to SAROPS; it will be able to predict not only where a search should be conducted, but when it should be stopped,” he added.

The U.S. Coast Guard turned to the University of Portsmouth researchers because of their expertise in the area of survival at sea, Turner said.

The model will be tested in late 2009 and may be rolled out across the Coast Guard system soon after, Tipton said.

A LEG UP

The path to a computer science degree can be more complicated for visually impaired students than for their sighted counterparts.

While some devices make it easier for visually impaired students to use computers, the visual nature of modern computing makes it difficult for them to participate in programming courses and in some other types of computing instruction, said Stephani Ludi, an associate professor of software engineering at the Rochester Institute of Technology.

Ludi is now principal investigator at Project Accessible Computing Education, an initiative funded by the National Science Foundation and headquartered at RIT.

The project aims to prepare visually impaired middle school and high school students for computer science programs at the college level. The ultimate goal is to make degrees in computer science accessible to greater numbers of visually impaired students, Ludi said.

“In the long run, encouraging these students will benefit everyone who uses a computer,” she said. “Because of the unique perspective they have, they can create software that is really more usable for everybody.”

ROBOTIC SERVICE DOG

Researchers at the Georgia Institute of Technology have engineered a robot that mirrors the actions of service dogs to help cut the costs and waiting times for real, live service dogs.

“It’s a road to get robots out there helping people sooner,” said Charlie Kemp, a professor of biomedical engineering at the school in Atlanta. “Service dogs have a great history of helping people, but there’s a multiyear waiting list. It’s a very expensive thing to have. We think robots will eventually help to meet those needs.”

To get the robotic dog to perform an action, users tell the robot to complete a task while aiming a laser pointer at the location of the action, Kemp said.

Kemp and graduate student Hai Nguyen worked closely with the team of trainers at Georgia Canines for Independence, or GCI, in Acworth, Ga., to come up with the command categories and interactions that are basic to the relationship between individuals and service dogs.

They primarily studied Betty, a golden retriever, to understand her movements and her relationship with commands. To open and close doors or drawers, for example, Betty pulls with her mouth on a towel attached to a drawer or door handle.

The robot was programmed to use the towel in a similar manner, Kemp said.

The robotic dog can replicate 10 tasks and commands taught to service dogs at GCI. Other successes included opening a microwave oven, delivering an object, and placing an item on a table, Kemp said.

Dog handlers at Georgia Canines for Independence were thrilled at the potential benefits of the technology.
“The waiting list for dogs can be five to seven years,” said Ramona Nichols, executive director of the agency. “It’s neat to see science happening, but with a bigger cause—applying the knowledge and experience we have, and really making a difference.”

A SMILING CAR

No one needs to tell Disney, which brought the likes of Herbie the Love Bug and Lightning McQueen to the big screen, that cars have personality.

Now a study co-authored by a Florida State University researcher has found through statistical analysis that many people see human facial features in the front ends of automobiles and ascribe various personality traits to cars.

“The study confirmed with some rigor what many people have already felt—that cars seem to have consistent personality traits associated with them and that this is similar to the way people perceive facial expressions,” said Dennis Slice, an associate professor in Florida State’s Department of Scientific Computing in Tallahassee.

The Disney Pixar movie Cars features the talking race car Lightning McQueen. A new study has found that many people ascribe human traits to automobiles.

As a guest professor at the University of Vienna, Slice collaborated with doctoral student Sonja Windhager, the study’s lead author, and several colleagues to explore the link between an automobile’s grille and headlights, and the way people perceive that car. The researchers asked 40 people to view high-resolution, 3-D computer reconstructions and printed images of 38 actual cars. The models represented cars produced by 26 automakers from 2004 through 2006.

One-third of the participants associated a human or animal face with at least 90 percent of the cars. Generally, the headlights were marked as eyes; the nose tended to be the grille or emblem; the additional air intake slots, the mouth, Slice said.

The participants also rated each model on 19 traits, including dominance, maturity, gender, and friendliness. They were also asked if they liked the car.

“In our study, people generally agreed in their ratings,” Slice said. “For instance, 96 percent of the participants agreed on whether a car was dominant or submissive.”

Cars scoring high in the so-called power traits had horizontally elongated hoods, pronounced lower car bodies relative to the windshields, and angular headlights that seemed to suggest a frown, Slice said. Cars on the other end of the power scale—those perceived as childlike, submissive, female, and friendly—had headlights with upper edges relatively close to the midline and had an upward shift of the car’s lateral-most points.

“In this way, the car gives us a big smile,” Slice said.

In a finding that suggests perhaps there is a hidden road warrior in all of us, study participants liked power vehicles best: the most mature, masculine, arrogant, and angry-looking ones, Slice said.

FASTEST COMPUTER

The petaflop IBM supercomputer nicknamed Roadrunner, which resides at Los Alamos National Laboratory in New Mexico, has held on to its top spot as the world’s fastest supercomputer, as ranked by an organization called Top500.

In May, the Los Alamos computer broke through the landmark barrier of one petaflop—or one quadrillion floating-point operations per second.

Following that feat, Roadrunner nabbed the No. 1 spot in June 2008. Fast on its heels is Jaguar, ranked as the second-fastest supercomputer by Top500. Jaguar is a Cray XT high-performance computing system at the Department of Energy’s Oak Ridge National Laboratory in Tennessee.

Twice a year, the Top500 list ranks the world’s top 500 supercomputers on their speed in running a benchmark program, High-Performance Linpack, or HPL. The list is compiled by Hans Meuer of the University of Mannheim, Germany; Erich Strohmaier and Horst Simon of the Lawrence Berkeley National Laboratory in California, and Jack Dongarra of the University of Tennessee in Knoxville.

Roadrunner posted a top performance of 1.059 petaflops in running the Linpack benchmark application. Although Roadrunner has been slightly enhanced since its initial win, it narrowly fended off a challenge by Jaguar, according to Meuer.

DETECTING A BAD MOOD

Who knows what evil lurks in the hearts of men? A computer might.

A researcher at Concordia University in Montreal is developing a computer image processing system that detects and classifies human facial expressions.

The system aims to analyze the photos it takes of individuals. It is for use in areas where security is a concern, such as an airport. If one could take random photos of the crowd and process them fast enough, there is the potential to identify individuals who might be problematic, said Prabir Bhattacharya, an information systems engineering professor at the university. He’s at work on the system with Concordia graduate student Abu Sayeed Sohail.

Facial expressions do not actually involve the entire face, but rather specific sets of muscles under the face near the eyes, nose and mouth, Bhattacharya said.

His system would measure 15 key points on the face and compare these measures against images of identifiable facial expressions. Although there is great variety in expression across individuals and cultures, the pair has identified seven basic expressions that seem to be relatively universal and has programmed the computer to recognize them, Bhattacharya said.

MOUNTAINOUS DESIGN

You’ve heard of customizing a design to fit the customer. When you are in the ski lift business, you customize the design to fit the mountain.

Inauen-Schaetti of Schwanden, Switzerland, designs cable systems, masts, and gondola stations for ski lifts, and inclined elevators for the transport systems used when drilling tunnels. Each system needs to be tailored to fit the mountain’s type, size, shape, terrain, and weather conditions, said Urs Thali, head of CAD design at the company.

The company recently brought in the CAD system CoCreate from PTC of Needham, Mass., that allows for radical engineering changes even late in the design process. That move helped slash project time by 15 percent, Thali said.

BRIEFLY NOTED

To help displaced engineers increase their marketability to employers, Altair Engineering of Troy, Mich., is offering free technology training courses and licenses for computer-aided engineering software. /// Dassault Systèmes of Paris is shipping CZone for Abaqus, from the company’s DS Simulia brand. The package is used to simulate the crushing of composite materials and is an add-on product for Abaqus FEA software. /// Autodesk Inc. of San Rafael, Calif., has signed a definitive agreement to acquire the iLogic software and related technology from Logimetrix Inc. of Windsor, Ontario. The software is desktop, rules-based, design automation technology. Terms were not disclosed. /// Coade Inc. of Houston has released an updated edition of its Coade Video Training Course for CADWorx Plant Professional. The course has been revised to take into account the latest functionality of AutoCAD 2009 and CADWorx Plant Professional 2009. /// Delcam of Birmingham, England, has released upgrades to its ArtCAM 2009 family of software for artistic applications. This includes the entry-level version, ArtCAM Express, for those new to CNC machining; ArtCAM Insignia for production machining, and ArtCAM Pro for more complex design and manufacturing. /// Theorem Solutions of Staffordshire, England, announced that its CADVerter now offers translation of CAD data between Catia version 5, revision 19, and the software programs NX, I-deas, CADDS5, ICEMsurf, and ProductView. /// Aftercad Software of Vancouver, British Columbia, has launched the commercial release of Aftercad Online, a service offering that allows businesses to view, mark up, manage, and publish large CAD files to the Web without the need for proprietary viewers. /// Nemetschek North America of Columbia, Md., has released Service Pack 1 for the Vectorworks 2009 product line. /// A maker of 3-D authoring software, ParallelGraphics of Dublin, Ireland, has released its Cortona3D Rapid Product range coupled with the Cortona3D viewer. The package gives companies the ability to create electronic spare parts catalogs, online interactive parts catalogs, and the like. /// A maker of digital manufacturing applications using the XVL format, Lattice Technology of San Francisco, has released three packages of its XVL software targeted at manufacturing functions. /// Mathematica 7 has been released by Wolfram Research of Champaign, Ill. /// Materialise of Leuven, Belgium, has upgraded Mimics, the company’s medical image processing package, to version 12.1.