Jeff's MCAD Blogging
Jeffrey Rowe has more than 40 years of experience in all aspects of industrial design, mechanical engineering, and manufacturing. On the publishing side, he has written well over 1,000 articles for CAD, CAM, CAE, and other technical publications, as well as consulting in many capacities in the … More »
October 19th, 2017 by Jeff Rowe
This week NVIDIA unveiled what it claims to be the world’s first artificial intelligence computer designed specifically to “drive” fully autonomous vehicles.
The new system, codenamed Pegasus, brings the NVIDIA® DRIVE™ PX AI computing platform for handling Level 5 driverless vehicles (Level 5 is ”steering wheel optional.” In other words, no human intervention is required, for example, a robotic taxi). NVIDIA DRIVE PX Pegasus can perform over 320 trillion operations per second — more than 10x the performance of its predecessor, NVIDIA DRIVE PX 2.
NVIDIA DRIVE PX Pegasus is intended to help make a new class of vehicles possible that can operate without a driver — fully autonomous vehicles without steering wheels, pedals, or mirrors, and interiors that feel more like a living room or office than a vehicle. They will arrive on demand to safely take passengers to their destinations, bringing mobility to everyone, including the elderly and disabled.
One of the driving forces behind autonomous vehicles is to recapture millions of hours of lost time that could be used by “drivers” (really passengers) to work, play, eat or sleep on their daily commutes. Theoretically, countless lives could be saved by vehicles that are never fatigued, impaired, or distracted — increasing road safety, reducing congestion, and possibly freeing up land currently used for parking lots.
Of the 225 partners developing on the NVIDIA DRIVE PX platform, more than 25 are developing fully autonomous robotaxis using NVIDIA CUDA GPUs. Today, their trunks resemble small data centers, loaded with racks of computers with server-class NVIDIA GPUs running deep learning, computer vision and parallel computing algorithms. Their size, power demands and cost make them impractical for production vehicles.
NVIDIA AI Vehicle Demonstration
The computational requirements of robotaxis are enormous — perceiving the world through high-resolution, 360-degree surround cameras and lidars, localizing the vehicle within centimeter accuracy, tracking vehicles and people around the car, and planning a safe and comfortable path to the destination. All this processing must be done with multiple levels of redundancy to ensure the highest level of safety. The computing demands of driverless vehicles are easily 50 to 100 times more intensive than the most advanced cars today with human drivers.
October 12th, 2017 by Jeff Rowe
This week PTC announced the upcoming release of the newest version of its Vuforia platform for AR development, Vuforia 7. The company says Vuforia 7 will introduce major advancements in the ability to attach digital content to more types of objects and environments, and a new capability for delivering enhanced AR experiences on a wide range of handheld devices.
PTC claims that Vuforia is the world’s most widely used platform for AR development. With support for mobile phones, tablets and eyewear, Vuforia has powered more than 475 million installs of AR apps from the App Store and Google Play. Vuforia Engine, the core of the platform, uses a device’s camera(s) and sensors to function as a digital “eye” inside an app. It “sees” objects and surfaces where content can be placed, and enables developers to create AR experiences using existing development tools.
Vuforia 7 will introduce Model Targets, a new feature for attaching content to objects that have not been recognizable using existing computer vision technology. Model Targets recognize objects by shape, in contrast to existing methods that rely on detailed visual designs typically found on print media, product packaging and many consumer goods. With Model Targets, content can be attached to objects such as automobiles, appliances, and industrial equipment and machinery. Model Targets will enable a new class of AR content that can replace traditional user manuals and technical service instructions.
Jay Wright, President, Vuforia: State of Augmented Reality
Responding to developers, Vuforia 7 will also introduce a new capability for placing content on horizontal surfaces. Vuforia Ground Plane enables content to be placed on the ground, floor or tabletop, whether indoors or outdoors. Vuforia Ground Plane extends the functionality of the Vuforia Smart Terrain feature, first announced in 2013 to take advantage of depth sensing cameras. Vuforia Ground Plane will support a wide range of today’s devices and provides an ideal solution for developers to build visualization apps, ranging from in-home furniture shopping to design review.
October 5th, 2017 by Jeff Rowe
It seems that a lot of CAD companies have taken a greater interest in digital simulation the past several years. Case in point — the recent MSC Software acquisition by Hexagon. That high level of interest was again evidenced this week as Dassault Systèmes and Exa Corporation with its simulation software for product engineering, announced the signing of a definitive merger agreement for Dassault Systèmes to acquire Exa. Under the terms of the merger agreement, this represents a value for Exa of approximately $400 million. Exa’s fiscal year ended January 31, 2017 and its revenue was $72 million.
With the addition of Exa, Dassault Systèmes’ 3DEXPERIENCE platform will provide customers with a mature, diverse portfolio of combined Lattice Boltzmann fluid simulation technologies, as well as Exa’s fully industrialized solutions and approximately 350 experienced simulation professionals. Because of its solving method, Exa’s solutions can solve fluids problems faster and more accurately than traditional methods for aerodynamics, aeroacoustics, and thermal management.
September 28th, 2017 by Jeff Rowe
A debate continues to rage about America’s general decline in science, engineering, and technology, largely blamed on the first steps of our youngest citizens – math and literacy education.
A couple of years ago I taught math to middle and high school students and witnessed firsthand not only the challenges, but also the opportunities for positive change brought about by a belief and commitment to teaching.
From my perspective, I learned that I taught to overcome two different but related needs – innumeracy (unfamiliarity with mathematical concepts and methods and the inability to use mathematics) as well as illiteracy (the inability to read and write). They both go hand in hand, because as important as getting the numbers right is, the ability to provide a convincing argument and communicate the numerical answer of the “why and how” is just as important.
September 21st, 2017 by Jeff Rowe
Interoperability, collaboration, inspection, quality, standards, proprietary data, neutrality, competition, and innovation. Over the years there have been myriad attempts to bring these processes together, all while protecting IP. However, as we know, while the attempts to make this happen have often been valiant, too often they have fallen well short, or worse, failed altogether.
That legacy of failure is on its way to being a thing of the past with the advent of the Quality Information Framework (QIF), an ANSI standard that supports digital thread concepts in engineering applications ranging from product design through manufacturing. Based on the XML standard, it contains a Library of XML Schema ensuring both data integrity and data interoperability in Model Based Enterprise (MBE) implementations.
QIF supports design, metrology, manufacturing, and is critical to the Industrial Revolution 4.0. Because it is XML based, QIF can be relatively easily integrated with Internet applications, and unlike other existing standards, there is no real barrier standing in the way for industry adopting QIF. It also effectively supports newer technologies, including additive manufacturing and the Internet of Things (IoT).
September 7th, 2017 by Jeff Rowe
Even though we’ve been told by a number of software vendors for several years now to use engineering simulation and analysis at the earliest stages of product development, relatively few companies have heeded the advice and actually done so. In many cases, it’s still design, break, repeat in a cycle that gets very expensive quickly trying to achieve optimized design goals. Even with all the insistence and chiding from the simulation folks, I’d estimate the percentage of design work that includes simulation early in the process as somewhere between 20-25%, although that may be a bit on the high side.
With it, engineers can rapidly explore design options and receive accurate simulation results with technology using engineering simulation to make digital exploration available to all engineers so they can design better products faster and more economically.
That’s a pretty confident and heady statement, knowing that several other vendors have attempted the roughly same thing with widely varying degrees of success. However, ANSYS has an interesting and innovative approach for reaching its goal — exploiting GPUs because they can handle massively parallel operations.
ANSYS readily admits that while Discovery Live is a means of bringing simulation to the engineering masses earlier in the development process, it doesn’t pretend to do everything for everybody, and there will always be a place for engineering simulation specialists for deeper dives. Discovery Live is targeted to early design exploration and to users new to simulation. Because it is not a solution for every simulation problem, Discovery Live does not compete with other more advanced ANSYS products, such as AIM, but data from it can be exported for more further study.
August 31st, 2017 by Jeff Rowe
A couple years ago I got into a pretty heated discussion with a staffer from an engineering software company about whether software patents were still relevant (or is they ever were to begin with).
While proponents (usually with deep pockets) have touted their benefits, software patents have also been used in the software industry to suppress innovation, kill competition, generate undeserved royalties, and make patent attorneys rich. So I’ll ask again, are software patents still relevant?
It’s no secret that the engineering software business is extremely competitive, as it always has been. Without naming names, the engineering software business has also proven to be a very fertile and lucrative ground for lawsuits regarding patent infringement, reverse engineering, and outright copying and pasting blocks of code.
Could stronger patent protection have prevented this from happening? Maybe yes, but probably, no.
Below is a video on the futility of software patents featuring Linus Torvalds, the creator, and for a long time, principal developer of the Linux kernel, which became the kernel for operating systems such as the Linux operating system, Android, and Chrome OS.
Linus Torvalds: Why Software Patents Make No Sense
Software patents have been hotly debated for years. Opponents to them have gained more visibility with less resources through the years than pro-patent supporters. Through these debates, arguments for and critiques against software patents have been focused mostly on the economic consequences of software patents, but there is a lot more to it than just money.
August 24th, 2017 by Jeff Rowe
Although the future of 3D printing continues to look bright, what is still needed is a new file format for 3D print data. Being very mindful of that fact, Autodesk, HP, Siemens, Stratasys, 3D Systems, and some others have come together to form the 3MF Consortium that espouses to get behind a truly ubiquitous file format for 3D printing. It’s really an industry partnership working toward the goal of finding a better, universally applicable 3D printing file format known as the 3D Manufacturing Format (3MF)—a file format originally developed by Microsoft, also a member of the Consortium.
The consortium admits that there is a problem that the 3D manufacturing must resolve – the current file formats used for 3D printing are in serious need of an upgrade. I totally agree.
Typically, data is passed from computer to 3D printer in STL (stereolithography) or OBJ (object) files, common 3D printing file formats. The 3MF Consortium, which now includes the research wing of General Electric, say STL and OBJ are outdated and clunky file formats with interoperability issues when used by some of the newer 3D printers, as well as contribute to 3D printing failures.
3MF Consortium Introduction
Thus, one of the driving forces behind 3MF, an XML-based open format, this new file type could contain information on the texture of a 3D print, the color of the print, and other complex characteristics. If that sounds familiar, that’s because it is—the Additive Manufacturing File Format (AMF), which has been around since 2011, solves many of the issues STL files have, and 3MF and AMF are in many respects pretty similar file formats, but let’s take a closer look.
August 17th, 2017 by Jeff Rowe
Like it or not, since the mid-1980s, the STL file format has been the de facto industry standard for transferring information between CAD programs and additive manufacturing equipment. However, the STL format only contains information about a surface mesh, and cannot represent color, texture, material, substructure, and other properties of a fabricated object.
As additive manufacturing technology has evolved from producing primarily single-material, homogenous shapes to producing multi-material geometries in full color with functionally graded materials and microstructures, there has been a growing need for a standard interchange file format that could support these features. A second factor that prompted the development of a new standard was the improving resolution of additive manufacturing machines. As the fidelity of printing processes approached micron scale resolution, the number of triangles required to describe smooth curved surfaces resulted in unacceptably large file sizes.
The Additive Manufacturing File Format (AMF) was introduced as an alternative to the STL file format to address many of the shortcomings of the popular file format. STL files introduce errors such as leaks and inconsistences, and also does not support color, material The choice, or orientation. STL files also rely on triangle subdivision to account for curvature. As the STL file scales in size, retaining resolution means introducing significantly more triangles. For example, a 10cm sphere at 10 micrometer resolution requires 20,000 triangles. Scaling up the 10cm sphere at the same resolution would significantly increase the amount of triangles, resulting in a much larger file. AMF seeks to address these issues by redesigning the way a 3D object is digitally stored.