Autodesk Research: Enabling the Future through Software



Autodesk researchers’ interests range from methods to help users learn powerful digital prototyping tools, to visualization and simulation techniques which enable designers to achieve new levels of performance.

Autodesk researchers’ interests range from methods to help users learn powerful digital prototyping tools, to visualization and simulation techniques which enable designers to achieve new levels of performance.

Autodesk — the company that “gives you the power to make anything” – is working to help solve some of the world’s most complex design problems, from pressing ecological challenges to the development of scalable smart infrastructure. Designers use Autodesk tools to not only create plans for buildings, for example, but also to simulate their impact on the environment and track their performance over time.

“Autodesk makes software for people who make things,” said Mark Davis, Senior Director, Autodesk Research, speaking at a recent Bloomberg panel discussion on Machine Learning, He said the company’s team of software developers and research scientists are “working into the future to figure out what the next set of tools software designers will need to design in this brave new world. Our current focus is on a technology that we call generative design.”

Autodesk is a huge company by any measure, with 170+ million users of consumer applications, 9,500+ employees worldwide, and products that are available in up to 17 languages. In 2016, total subscriptions increased by approximately 345,000 to 2.58 million. Autodesk also provides free access to its software to students, teachers and academic institutions worldwide.

Most customers interface with the Autodesk Account part of the company to license their subscriptions and get updates, but commercially available products are first born, tested and refined in the company’s research, labs and beta arms (Figure 1).

According to Scott Shepard, a Program Manager and blogger for Autodesk Labs, the Autodesk Research team works with universities on cutting edge research. They are free thinkers who can propose any idea that is “so crazy, it just might work.” That leads to innovations that would otherwise be unheard of.

Autodesk Labs was created to handle questions such as “What if we totally redid this?” or “What if we made it do something else instead?” When something is on Labs, it is early enough in the life cycle that questions like this are welcome.

Prior to release, some customers participate in Autodesk beta programs. These are typically done under non-disclosure agreements (NDA). The focus of a beta is to make sure a product is ready to ship.

“Ideally, an idea would start out in Autodesk Research, get a thumbs up from early adopters as part of an Autodesk Labs technology preview, get fully developed by the product development organization, get beta tested, and released as part of Autodesk’s subscription offerings,” Shepard explained in a recent blog.

Autodesk-F1

Figure 1: Commercially available products are born, tested and refined in Autodesk’s research, labs and beta groups.

To provide an overview of where Autodesk is heading in the next few years, this article looks at the various groups that make up the Autodesk Research part of the company and highlights some of their current projects. Autodesk researchers’ interests range from methods to help users learn powerful digital prototyping tools, to visualization and simulation techniques which enable designers to achieve new levels of performance. The groups, which collaborate openly with researchers at leading universities around the world, are focused on: Bio/Nano, Complex Systems, Computational Science, Design and Social Impact, Machine Intelligence, Simulation and Graphics, Design, Design and Fabrication, User Interfaces, and Research Transfer.

Bio/Nano Research

The aim of the Autodesk bio/nano research group collaborates with researchers to co-envision the paradigms and tools needed to understand and exploit the intersection of design with life & materials sciences. The group also explore and drives the emergent design spaces enabled by bio/nano such as synthetic biology.

“For thirty years, we’ve been creating software to help people create and design all the things around us,” said Florencio Mazzoldi, head of engineering, bio/nano research group. “We’re taking some of that software and some of that knowledge around design and creation and manufacturing and bringing it into the new domains of biology and nanotechnology.”

One of the group’s current projects is a Molecule Viewer, a web-based 3D tool optimized for viewing, exploring, and sharing of large-scale protein datasets from the RCSB Protein Data Bank directly in the web browser (Figure 2).

Figure 2: A 3.6-Angstrom cryoEM structure of human adenovirus type 5, made up of 6 million atoms, shown in the web browser on Autodesk Molecule Viewer.

Figure 2: A 3.6-Angstrom cryoEM structure of human adenovirus type 5, made up of 6 million atoms, shown in the web browser on Autodesk Molecule Viewer.

3D visualization of large and multi-scale biological data, from macro-molecular structures to whole organisms, is integral to building models for biomedical research. Existing standards for desktop, plug-in, and web-based applications for 3D visualization can easily push beyond the limits of advanced processing and rendering.

The Autodesk Research Molecule Viewer and its web-based 3D visualization platform is built to leverage cloud capabilities to overcome limitations of scalability, capability, accessibility, collaboration, and communication for 3D biological datasets.

Beyond the Molecule Viewer application, the group says this effort will serve as a stepping stone to launch investigations extending this 3D visualization framework to other large-scale scientific 3D data, making exploration and design of large 3D data scalable, accessible, and shareable.

The group’s current projects also include bioprinting, a genetic constructor and nanodesign. The bioprinting project has two general objectives: the development and adaptation of biological materials to make them usable in 3D printers, and the use of these and other materials to further research in tissue engineering and regenerative medicine. Genetic Constructor is an extensible, open source, cloud CAD tool to drive biological design and complex DNA construction. Project NanoDesign aims to create the new generation of CAD tools to enable the design and manufacturing of nanoscale structures with the goal to accelerate the adoption of DNA nanotechnology and continue to push boundaries in research and application.

Complex Systems Research

The goal of the complex systems research group is to help designers and researchers gain a deeper understanding of complex systems such as those found in the biological world and the built environment. Specifically, the group develops tools and techniques to model, simulate, and explore human-centric natural and designed systems, with an emphasis on data collection and visualization. The team strives to provide a novel multi-paradigm, multi-scale, and multi-physics platform to support collaboration among experts in a multitude of domains.

In the Human Systems Biology domain, for example, the group partners with experts in organizations, such as the Parametric Human Project, to develop novel multi-scale simulation models. In the domain of the built environment, the group partners with academia and government laboratories in Systems Design for Sustainability.

One of the group’s main projects is Systems Design and Simulation. While traditional programming practices have produced a wide range of relatively independent simulation methods, predictive models of extremely complex natural and artificial systems will require a more scalable, more collaborative approach to modeling. This project strives for software that will help researchers develop, debug, document, share, and integrate simulation code.

Another project is focused on Computational Anatomy and Biomechanics, the goal of which is to understand not only the range of external shapes of individual anatomical objects (e.g., muscles, bones), but also their internal architectures, as well as mapping the connections and spatial relations amongst anatomical elements (Figure 3). The group is working to develop a “visual grammar” to describe the surface and volumetric features of organic objects, for the purpose of creating ontological models of the human anatomy that can be understood by both man and machine.

Figure 3: The computational anatomy and biomechanics project seeks to understand not only the range of external shapes of individual anatomical objects (e.g., muscles, bones), but also their internal architectures.

Figure 3: The computational anatomy and biomechanics project seeks to understand not only the range of external shapes of individual anatomical objects (e.g., muscles, bones), but also their internal architectures.

Multiscale Interaction is another of the group’s projects. Computers and 3D graphics applications are continuously increasing in power, memory, and rendering capabilities, making larger and more complex 3D scenes possible. Domains such as medical visualization, architecture and urban design, geospatial scanning, astrophysics, biochemistry, and abstract data analysis are beginning to consider massive datasets. Many of these datasets contain objects that exist at multiple scales, that is, the objects have meaningful observable properties at scales that are one or more orders of magnitude apart. As the complexity of datasets continues to grow, multiscale approaches to interactions are critical to keeping the user in the loop.

The group also has a Visualization & Visual Analytics project underway to tackle the question, “what makes a visualization effective?” Another project is called Dasher uses a BIM (Building Information Modeling)-based platform to provide building owners with greater insight into real-time building performance throughout the life-cycle of the building. Ramtin Attar, a design research associate with Autodesk, explained the goal: “Today we still face a great disconnect between our design tools and how we design our buildings and how we operate them. We need to move beyond fixed benchmarks and look at measuring sustainability and ‘greenness.’ This is as much about providing better tools to bridge this gap and better sustainability as it is about changing the mindset of how we look at buildings, not as static things but as ecosystems.”

Design Research

The Design research group is focused on discovering new, innovative user experiences for designers. Generally taking the form of interactive prototypes and simulations, the group’s research incorporates aspects of ethnography, cognitive science, human-computer interaction, usability, and other design and user research related domains. The goal is to create and refine innovative cloud and mobile experiences that are easily used by consumers, while still scaling up to the demands of professional design and engineering customers.

Current areas of interest include advances in multitouch and gestural interfaces, leveraging the cloud to facilitate design “optioneering”, and rapid application delivery into mobile environments.

The group’s Project Saturn is focused on cloud-based computing and optimization. Mathematical optimization techniques often require immense computational power to perform in a timely manner.  The advent of large-scale cloud computing infrastructures makes it possible to finally solve large optimization problems. Project Saturn, which is also directed by the Computational Science Research Group, aims to develop such a generic cloud-based optimization framework.

Specifically, the Saturn framework is designed to provide a complete library of single- and multi-objective global optimization algorithms, which can be utilized both on local computing resources as well as on the cloud. Projects can integrate Saturn directly as a multi-language multi-platform optimization library with a dedicated API or through a web-service based API to communicate with the global optimization framework running inside the Autodesk Cloud. Results can be seen through Saturn’s own advanced visualization toolkit, which is available through its web interface.

Project Dreamcatcher, which employs the Saturn framework, is described as the next generation of CAD. It is a generative design system that enables designers to craft a definition of their design problem through goals and constraints. This information is used to synthesize alternative design solutions that meet the objectives. Designers are able to explore tradeoffs between many alternative approaches and select design solutions for manufacture.

The Dreamcatcher system allows designers to input specific design objectives, including functional requirements, material type, manufacturing method, performance criteria, and cost restrictions. Loaded with design requirements, the system then searches a procedurally synthesized design space to evaluate a vast number of generated designs for satisfying the design requirements. The resulting design alternatives are then presented back to the user, along with the performance data of each solution, in the context of the entire design solution space. Designers are able to evaluate the generated solutions in real time, returning at any point to the problem definition to adjust goals and constraints to generate new results that fit the refined definition of success. Once the design space has been explored to satisfaction, the designer is able to output the design to fabrication tools or export the resulting geometry for use in other software tools.

Design and Fabrication Group

The Design and Fabrication group explores new ways to create objects in both the virtual and physical domains. Generally taking the form of prototype interactive systems, the research incorporates aspects of computer graphics, human-computer interaction, interactive systems, and other related domains. The goal is to create novel design interfaces which are immediately accessible to a novice, while still scaling up to the demands of a professional.

One of the group’s current projects is Meshmixer, a prototype design tool based on high-resolution dynamic triangle meshes. Another related project is Mesh Processing where novel approaches to geometry processing are explored.

Machine Intelligence

Our Machine Intelligence team works in an iterative fashion to research and develop new techniques to apply modern machine learning to designing, creating and interpreting the world. Expectations on the complexity of designs are rapidly increasing and with new manufacturing and materials techniques entering the market new methods are required to assist designers in realizing effective designs.

Not only are sensors becoming cheap and ubiquitous, allowing constant monitoring of the world, but design information is also rich in semantic content.  By analyzing and learning from this information, sophisticated models can be built to: accelerate and elevate the design process, explore entirely new design concepts, automatically direct design based on real performance feedback, and assist with the construction or assembly of products. The group works in a number of areas including: Deep learning, reinforcement learning, knowledge modeling and representation, geometric shape analysis, and robotic control and sensing.

Simulation and Graphics

The Simulation and Graphics group has a long history of developing advanced natural phenomena simulation. The team’s research has resulted in a real-time fluids solver, an advanced cloth solver and self-affecting particle system.

Currently, the major focus of the group is on expanding and improving its unified solver for dynamics, called Nucleus. Existing dynamics solvers are designed for specific effects such as rigid bodies or cloth. Handling the interaction between these solvers is often problematic as one of them takes precedence over the others. The team’s approach is to evaluate all constraints together, resulting in more stable results. The basic philosophy behind Nucleus is that complexity arises by combining simple constraints.

User Interface Research

The User Interface Research Group is focused on advancing the field of human-computer interaction research both at the hardware and software levels. The team is exploring new ways to make software easier to learn, friendlier to use, and more efficient. Interests lie in both seeking novel yet immediately applicable solutions to the user interface challenges existing in today’s software applications, and also rethinking the underlying concepts for which today’s user interfaces are designed, by exploring new paradigms which could shape the future generation of application user interfaces.

The group’s work spans multiple areas within the human-computer interaction research field, including interaction design, computer supported collaborative work, information visualization, and 3D user interfaces.

Share and Enjoy:
  • Digg
  • Sphinn
  • del.icio.us
  • Facebook
  • Mixx
  • Google
  • TwitThis

Tags:

  • Free Download

    na_cover_98

    Download the latest issue of the Autodesk® Partner Solutions Design Guide 2018 and subscribe to receive future issues and the quarterly email newsletter