Category: Visual Design

Want to design the car of the future? Here are 8,000 designs to get you started.

Car design is an iterative and proprietary process. Carmakers can spend several years on the design phase for a car, tweaking 3D forms in simulations before building out the most promising designs for physical testing. The details and specs of these tests, including the aerodynamics of a given car design, are typically not made public. Significant advances in performance, such as in fuel efficiency or electric vehicle range, can therefore be slow and siloed from company to company.

MIT engineers say that the search for better car designs can speed up exponentially with the use of generative artificial intelligence tools that can plow through huge amounts of data in seconds and find connections to generate a novel design. While such AI tools exist, the data they would need to learn from have not been available, at least in any sort of accessible, centralized form.

But now, the engineers have made just such a dataset available to the public for the first time. Dubbed DrivAerNet++, the dataset encompasses more than 8,000 car designs, which the engineers generated based on the most common types of cars in the world today. Each design is represented in 3D form and includes information on the car’s aerodynamics — the way air would flow around a given design, based on simulations of fluid dynamics that the group carried out for each design.

In a new dataset that includes more than 8,000 car designs, MIT engineers simulate the aerodynamics for a given car shape, which they represent in various modalities, including “surface fields” (left) and “streamlines” (right).

Credit: Courtesy of Mohamed Elrefaie

Each of the dataset’s 8,000 designs is available in several representations, such as mesh, point cloud, or a simple list of the design’s parameters and dimensions. As such, the dataset can be used by different AI models that are tuned to process data in a particular modality.

DrivAerNet++ is the largest open-source dataset for car aerodynamics that has been developed to date. The engineers envision it being used as an extensive library of realistic car designs, with detailed aerodynamics data that can be used to quickly train any AI model. These models can then just as quickly generate novel designs that could potentially lead to more fuel-efficient cars and electric vehicles with longer range, in a fraction of the time that it takes the automotive industry today.

“This dataset lays the foundation for the next generation of AI applications in engineering, promoting efficient design processes, cutting R&D costs, and driving advancements toward a more sustainable automotive future,” says Mohamed Elrefaie, a mechanical engineering graduate student at MIT.

Elrefaie and his colleagues will present a paper detailing the new dataset, and AI methods that could be applied to it, at the NeurIPS conference in December. His co-authors are Faez Ahmed, assistant professor of mechanical engineering at MIT, along with Angela Dai, associate professor of computer science at the Technical University of Munich, and Florin Marar of BETA CAE Systems.

Filling the data gap

Ahmed leads the Design Computation and Digital Engineering Lab (DeCoDE) at MIT, where his group explores ways in which AI and machine-learning tools can be used to enhance the design of complex engineering systems and products, including car technology.

“Often when designing a car, the forward process is so expensive that manufacturers can only tweak a car a little bit from one version to the next,” Ahmed says. “But if you have larger datasets where you know the performance of each design, now you can train machine-learning models to iterate fast so you are more likely to get a better design.”

And speed, particularly for advancing car technology, is particularly pressing now.

“This is the best time for accelerating car innovations, as automobiles are one of the largest polluters in the world, and the faster we can shave off that contribution, the more we can help the climate,” Elrefaie says.

In looking at the process of new car design, the researchers found that, while there are AI models that could crank through many car designs to generate optimal designs, the car data that is actually available is limited. Some researchers had previously assembled small datasets of simulated car designs, while car manufacturers rarely release the specs of the actual designs they explore, test, and ultimately manufacture.

The team sought to fill the data gap, particularly with respect to a car’s aerodynamics, which plays a key role in setting the range of an electric vehicle, and the fuel efficiency of an internal combustion engine. The challenge, they realized, was in assembling a dataset of thousands of car designs, each of which is physically accurate in their function and form, without the benefit of physically testing and measuring their performance.

To build a dataset of car designs with physically accurate representations of their aerodynamics, the researchers started with several baseline 3D models that were provided by Audi and BMW in 2014. These models represent three major categories of passenger cars: fastback (sedans with a sloped back end), notchback (sedans or coupes with a slight dip in their rear profile) and estateback (such as station wagons with more blunt, flat backs). The baseline models are thought to bridge the gap between simple designs and more complicated proprietary designs, and have been used by other groups as a starting point for exploring new car designs.

Library of cars

In their new study, the team applied a morphing operation to each of the baseline car models. This operation systematically made a slight change to each of 26 parameters in a given car design, such as its length, underbody features, windshield slope, and wheel tread, which it then labeled as a distinct car design, which was then added to the growing dataset. Meanwhile, the team ran an optimization algorithm to ensure that each new design was indeed distinct, and not a copy of an already-generated design. They then translated each 3D design into different modalities, such that a given design can be represented as a mesh, a point cloud, or a list of dimensions and specs.

The researchers also ran complex, computational fluid dynamics simulations to calculate how air would flow around each generated car design. In the end, this effort produced more than 8,000 distinct, physically accurate 3D car forms, encompassing the most common types of passenger cars on the road today.

To produce this comprehensive dataset, the researchers spent over 3 million CPU hours using the MIT SuperCloud, and generated 39 terabytes of data. (For comparison, it’s estimated that the entire printed collection of the Library of Congress would amount to about 10 terabytes of data.)

The engineers say that researchers can now use the dataset to train a particular AI model. For instance, an AI model could be trained on a part of the dataset to learn car configurations that have certain desirable aerodynamics. Within seconds, the model could then generate a new car design with optimized aerodynamics, based on what it has learned from the dataset’s thousands of physically accurate designs.

The researchers say the dataset could also be used for the inverse goal. For instance, after training an AI model on the dataset, designers could feed the model a specific car design and have it quickly estimate the design’s aerodynamics, which can then be used to compute the car’s potential fuel efficiency or electric range — all without carrying out expensive building and testing of a physical car.

“What this dataset allows you to do is train generative AI models to do things in seconds rather than hours,” Ahmed says. “These models can help lower fuel consumption for internal combustion vehicles and increase the range of electric cars — ultimately paving the way for more sustainable, environmentally friendly vehicles.”

This work was supported, in part, by the German Academic Exchange Service and the Department of Mechanical Engineering at MIT.

The Law of Diminishing Returns

Striking the right balance can be tough. We don’t want cool mama bear’s porridge or hot papa’s bear porridge, but something right in the middle, like baby bear’s porridge.

The Law of Diminishing Returns originally published on CSS-Tricks, which is part of the DigitalOcean family. You should…

One of Those “Onboarding” UIs, With Anchor Positioning

We can anchor one element to another. We can also attach one element to multiple anchors. In this experiment, Ryan riffs on those ideas and comes up with a new way to transition between two anchors and the result is a practical use case that would…

WordPress Multi-Multisite: A Case Study

What’s it look like to create a dashboard within the WordPress admin for analyzing Google Analytics data across 900 blogs across 25 multisite instances? It involves designing a user-friendly interface, leveraging the WordPress REST API, implementing a plugin for data retrieval, and addressing challenges like rate…

To design better water filters, MIT engineers look to manta rays

Filter feeders are everywhere in the animal world, from tiny crustaceans and certain types of coral and krill, to various molluscs, barnacles, and even massive basking sharks and baleen whales. Now, MIT engineers have found that one filter feeder has evolved to sift food in ways that could improve the design of industrial water filters.

In a paper appearing this week in the Proceedings of the National Academy of Sciences, the team characterizes the filter-feeding mechanism of the mobula ray — a family of aquatic rays that includes two manta species and seven devil rays. Mobula rays feed by swimming open-mouthed through plankton-rich regions of the ocean and filtering plankton particles into their gullet as water streams into their mouths and out through their gills.

The floor of the mobula ray’s mouth is lined on either side with parallel, comb-like structures, called plates, that siphon water into the ray’s gills. The MIT team has shown that the dimensions of these plates may allow for incoming plankton to bounce all the way across the plates and further into the ray’s cavity, rather than out through the gills. What’s more, the ray’s gills absorb oxygen from the outflowing water, helping the ray to simultaneously breathe while feeding.

“We show that the mobula ray has evolved the geometry of these plates to be the perfect size to balance feeding and breathing,” says study author Anette “Peko” Hosoi, the Pappalardo Professor of Mechanical Engineering at MIT.

The engineers fabricated a simple water filter modeled after the mobula ray’s plankton-filtering features. They studied how water flowed through the filter when it was fitted with 3D-printed plate-like structures. The team took the results of these experiments and drew up a blueprint, which they say designers can use to optimize industrial cross-flow filters, which are broadly similar in configuration to that of the mobula ray.

“We want to expand the design space of traditional cross-flow filtration with new knowledge from the manta ray,” says lead author and MIT postdoc Xinyu Mao PhD ’24. “People can choose a parameter regime of the mobula ray so they could potentially improve overall filter performance.”

Hosoi and Mao co-authored the new study with Irmgard Bischofberger, associate professor of mechanical engineering at MIT.

A better trade-off

The new study grew out of the group’s focus on filtration during the height of the Covid pandemic, when the researchers were designing face masks to filter out the virus. Since then, Mao has shifted focus to study filtration in animals and how certain filter-feeding mechanisms might improve filters used in industry, such as in water treatment plants.

Mao observed that any industrial filter must strike a balance between permeability (how easily fluid can flow through a filter), and selectivity (how successful a filter is at keeping out particles of a target size). For instance, a membrane that is studded with large holes might be highly permeable, meaning a lot of water can be pumped through using very little energy. However, the membrane’s large holes would let many particles through, making it very low in selectivity. Likewise, a membrane with much smaller pores would be more selective yet also require more energy to pump the water through the smaller openings.

“We asked ourselves, how do we do better with this tradeoff between permeability and selectivity?” Hosoi says.

As Mao looked into filter-feeding animals, he found that the mobula ray has struck an ideal balance between permeability and selectivity: The ray is highly permeable, in that it can let water into its mouth and out through its gills quickly enough to capture oxygen to breathe. At the same time, it is highly selective, filtering and feeding on plankton rather than letting the particles stream out through the gills.

The researchers realized that the ray’s filtering features are broadly similar to that of industrial cross-flow filters. These filters are designed such that fluid flows across a permeable membrane that lets through most of the fluid, while any polluting particles continue flowing across the membrane and eventually out into a reservoir of waste.

The team wondered whether the mobula ray might inspire design improvements to industrial cross-flow filters. For that, they took a deeper dive into the dynamics of mobula ray filtration.

A vortex key

As part of their new study, the team fabricated a simple filter inspired by the mobula ray. The filter’s design is what engineers refer to as a “leaky channel” — effectively, a pipe with holes along its sides. In this case, the team’s “channel” consists of two flat, transparent acrylic plates that are glued together at the edges, with a slight opening between the plates through which fluid can be pumped. At one end of the channel, the researchers inserted 3D-printed structures resembling the grooved plates that run along the floor of the mobula ray’s mouth.

The team then pumped water through the channel at various rates, along with colored dye to visualize the flow. They took images across the channel and observed an interesting transition: At slow pumping rates, the flow was “very peaceful,” and fluid easily slipped through the grooves in the printed plates and out into a reservoir. When the researchers increased the pumping rate, the faster-flowing fluid did not slip through, but appeared to swirl at the mouth of each groove, creating a vortex, similar to a small knot of hair between the tips of a comb’s teeth.

“This vortex is not blocking water, but it is blocking particles,” Hosoi explains. “Whereas in a slower flow, particles go through the filter with the water, at higher flow rates, particles try to get through the filter but are blocked by this vortex and are shot down the channel instead. The vortex is helpful because it prevents particles from flowing out.”

The team surmised that vortices are the key to mobula rays’ filter-feeding ability. The ray is able to swim at just the right speed that water, streaming into its mouth, can form vortices between the grooved plates. These vortices effectively block any plankton particles — even those that are smaller than the space between plates. The particles then bounce across the plates and head further into the ray’s cavity, while the rest of the water can still flow between the plates and out through the gills.

The researchers used the results of their experiments, along with dimensions of the filtering features of mobula rays, to develop a blueprint for cross-flow filtration.

“We have provided practical guidance on how to actually filter as the mobula ray does,” Mao offers.

“You want to design a filter such that you’re in the regime where you generate vortices,” Hosoi says. “Our guidelines tell you: If you want your plant to pump at a certain rate, then your filter has to have a particular pore diameter and spacing to generate vortices that will filter out particles of this size. The mobula ray is giving us a really nice rule of thumb for rational design.”

This work was supported, in part, by the U.S. National Institutes of Health, and the Harvey P. Greenspan Fellowship Fund.

Alt Text: Not Always Needed

Most images require description for clarity, there are exceptions. This set of notes looks at different situations and contexts where alt text may not be needed and what to do in those cases.

Alt Text: Not Always Needed originally published on CSS-Tricks, which is part of the…

Solved by CSS: Donuts Scopes

Donut scoping addresses the challenge of preventing parent styles from leaking to nested content. Originating from a 2011 concept by Nicole Sullivan, the issue has evolved, culminating in 2024’s @scope at-rule. This allows for more precise CSS styling, safeguarding content from unwanted inheritance while managing global…

Invoker Commands: Additional Ways to Work With Dialog, Popover… and More?

Web browsers are experimenting with two HTML attributes — technically, they’re called “invoker commands” — that are designed to invoke popovers, dialogs, and further down the line, all kinds of actions without writing JavaScript. Although, if you do reach for JavaScript, the new attributes come with…

Dancing with currents and waves in the Maldives

Any child who’s spent a morning building sandcastles only to watch the afternoon tide ruin them in minutes knows the ocean always wins.

Yet, coastal protection strategies have historically focused on battling the sea — attempting to hold back tides and fighting waves and currents by armoring coastlines with jetties and seawalls and taking sand from the ocean floor to “renourish” beaches. These approaches are temporary fixes, but eventually the sea retakes dredged sand, intense surf breaches seawalls, and jetties may just push erosion to a neighboring beach. The ocean wins.

With climate change accelerating sea level rise and coastal erosion, the need for better solutions is urgent. Noting that eight of the world’s 10 largest cities are near a coast, a recent National Oceanic and Atmospheric Administration (NOAA) report pointed to 2023’s record-high global sea level and warned that high tide flooding is now 300 to 900 percent more frequent than it was 50 years ago, threatening homes, businesses, roads and bridges, and a range of public infrastructure, from water supplies to power plants.

Island nations face these threats more acutely than other countries and there’s a critical need for better solutions. MIT’s Self-Assembly Lab is refining an innovative one that demonstrates the value of letting nature take its course — with some human coaxing.

The Maldives, an Indian Ocean archipelago of nearly 1,200 islands, has traditionally relied on land reclamation via dredging to replenish its eroding coastlines. Working with the Maldivian climate technology company Invena Private Limited, the Self-Assembly Lab is pursuing technological solutions to coastal erosion that mimic nature by harnessing ocean currents to accumulate sand. The Growing Islands project creates and deploys underwater structures that take advantage of wave energy to promote accumulation of sand in strategic locations — helping to expand islands and rebuild coastlines in sustainable ways that can eventually be scaled to coastal areas around the world.

“There’s room for a new perspective on climate adaptation, one that builds with nature and leverages data for equitable decision-making,” says Invena co-founder and CEO Sarah Dole.

MIT’s pioneering work was the topic of multiple presentations during the United Nations General Assembly and Climate week in New York City in late September. During the week, Self-Assembly Lab co-founder and director Skylar Tibbits and Maldives Minister of Climate Change, Environment and Energy Thoriq Ibrahim also presented findings of the Growing Islands project at MIT Solve’s Global Challenge Finals in New York.

“There’s this interesting story that’s emerging around the dynamics of islands,” says Tibbits, whose U.N.-sponsored panel (“Adaptation Through Innovation: How the Private Sector Could Lead the Way”) was co-hosted by the Government of Maldives and the U.S. Agency for International Development, a Growing Islands project funder.

In a recent interview, Tibbits said islands “are almost lifelike in their characteristics. They can adapt and grow and change and fluctuate.” Despite some predictions that the Maldives might be inundated by sea level rise and ravaged by erosion, “maybe these islands are actually more resilient than we thought. And maybe there’s a lot more we can learn from these natural formations of sand … maybe they are a better model for how we adapt in the future for sea level rise and erosion and climate change than our man-made cities.”

Building on a series of lab experiments begun in 2017, the MIT Self-Assembly Lab and Invena have been testing the efficacy of submersible structures to expand islands and rebuild coasts in the Maldivian capital of Male since 2019. Since then, researchers have honed the experiments based on initial results that demonstrate the promise of using submersible bladders and other structures to utilize natural currents to encourage strategic accumulation of sand.

The work is “boundary-pushing,” says Alex Moen, chief explorer engagement officer at the National Geographic Society, an early funder of the project.

“Skylar and his team’s innovative technology reflect the type of forward-thinking, solutions-oriented approaches necessary to address the growing threat of sea level rise and erosion to island nations and coastal regions,” Moen said.

Most recently, in August 2024, the team submerged a 60-by-60-meter structure in a lagoon near Male. The structure is six times the size of its predecessor installed in 2019, Tibbits says, adding that while the 2019 island-building experiment was a success, ocean currents in the Maldives change seasonally and it only allowed for accretion of sand in one season.

“The idea of this was to make it omnidirectional. We wanted to make it work year-round. In any direction, any season, we should be accumulating sand in the same area,” Tibbits says. “This is our largest experiment so far, and I think it has the best chance to accumulate the most amount of sand, so we’re super excited about that.”

The next experiment will focus not on building islands, but on overcoming beach erosion. This project, planned for installation later this fall, is envisioned to not only enlarge a beach but also provide recreational benefits for local residents and enhanced habitat for marine life such as fish and corals.

“This will be the first large-scale installment that’s intentionally designed for marine habitats,” Tibbits says.

Another key aspect of the Growing Islands project takes place in Tibbits’ lab at MIT, where researchers are improving the ability to predict and track changes in low-lying islands through satellite imagery analysis — a technique that promises to facilitate what is now a labor-intensive process involving land and sea surveys by drones and researchers on foot and at sea.

“In the future, we could be monitoring and predicting coastlines around the world — every island, every coastline around the world,” Tibbits says. “Are these islands getting smaller, getting bigger? How fast are they losing ground? No one really knows unless we do it by physically surveying right now and that’s not scalable. We do think we have a solution for that coming.”

Also hopefully coming soon is financial support for a Mobile Ocean Innovation Lab, a “floating hub” that would provide small island developing states with advanced technologies to foster coastal and climate resilience, conservation, and renewable energy. Eventually, Tibbits says, it would enable the team to travel “any place around the world and partner with local communities, local innovators, artists, and scientists to help co-develop and deploy some of these technologies in a better way.”

Expanding the reach of climate change solutions that collaborate with, rather than oppose, natural forces depends on getting more people, organizations, and governments on board.

“There are two challenges,” Tibbits says. “One of them is the legacy and history of what humans have done in the past that constrains what we think we can do in the future. For centuries, we’ve been building hard infrastructure at our coastlines, so we have a lot of knowledge about that. We have companies and practices and expertise, and we have a built-up confidence, or ego, around what’s possible. We need to change that.

“The second problem,” he continues, “is the money-speed-convenience problem — or the known-versus-unknown problem. The hard infrastructure, whether that’s groins or seawalls or just dredging … these practices in some ways have a clear cost and timeline, and we are used to operating in that mindset. And nature doesn’t work that way. Things grow, change, and adapt on their on their own timeline.”

Teaming up with waves and currents to preserve islands and coastlines requires a mindset shift that’s difficult, but ultimately worthwhile, Tibbits contends.

“We need to dance with nature. We’re never going to win if we’re trying to resist it,” he says. “But the best-case scenario is that we can take all the positive attributes in the environment and take all the creative, positive things we can do as humans and work together to create something that’s more than the sum of its parts.”

Graph-based AI model maps the future of innovation

Imagine using artificial intelligence to compare two seemingly unrelated creations — biological tissue and Beethoven’s “Symphony No. 9.” At first glance, a living system and a musical masterpiece might appear to have no connection. However, a novel AI method developed by Markus J. Buehler, the McAfee Professor of Engineering and professor of civil and environmental engineering and mechanical engineering at MIT, bridges this gap, uncovering shared patterns of complexity and order.

“By blending generative AI with graph-based computational tools, this approach reveals entirely new ideas, concepts, and designs that were previously unimaginable. We can accelerate scientific discovery by teaching generative AI to make novel predictions about never-before-seen ideas, concepts, and designs,” says Buehler.

The open-access research, recently published in Machine Learning: Science and Technology, demonstrates an advanced AI method that integrates generative knowledge extraction, graph-based representation, and multimodal intelligent graph reasoning.

The work uses graphs developed using methods inspired by category theory as a central mechanism to teach the model to understand symbolic relationships in science. Category theory, a branch of mathematics that deals with abstract structures and relationships between them, provides a framework for understanding and unifying diverse systems through a focus on objects and their interactions, rather than their specific content. In category theory, systems are viewed in terms of objects (which could be anything, from numbers to more abstract entities like structures or processes) and morphisms (arrows or functions that define the relationships between these objects). By using this approach, Buehler was able to teach the AI model to systematically reason over complex scientific concepts and behaviors. The symbolic relationships introduced through morphisms make it clear that the AI isn’t simply drawing analogies, but is engaging in deeper reasoning that maps abstract structures across different domains.

Buehler used this new method to analyze a collection of 1,000 scientific papers about biological materials and turned them into a knowledge map in the form of a graph. The graph revealed how different pieces of information are connected and was able to find groups of related ideas and key points that link many concepts together.

“What’s really interesting is that the graph follows a scale-free nature, is highly connected, and can be used effectively for graph reasoning,” says Buehler. “In other words, we teach AI systems to think about graph-based data to help them build better world representations models and to enhance the ability to think and explore new ideas to enable discovery.”

Researchers can use this framework to answer complex questions, find gaps in current knowledge, suggest new designs for materials, and predict how materials might behave, and link concepts that had never been connected before.

The AI model found unexpected similarities between biological materials and “Symphony No. 9,” suggesting that both follow patterns of complexity. “Similar to how cells in biological materials interact in complex but organized ways to perform a function, Beethoven’s 9th symphony arranges musical notes and themes to create a complex but coherent musical experience,” says Buehler.

In another experiment, the graph-based AI model recommended creating a new biological material inspired by the abstract patterns found in Wassily Kandinsky’s painting, “Composition VII.” The AI suggested a new mycelium-based composite material. “The result of this material combines an innovative set of concepts that include a balance of chaos and order, adjustable property, porosity, mechanical strength, and complex patterned chemical functionality,” Buehler notes. By drawing inspiration from an abstract painting, the AI created a material that balances being strong and functional, while also being adaptable and capable of performing different roles. The application could lead to the development of innovative sustainable building materials, biodegradable alternatives to plastics, wearable technology, and even biomedical devices.

With this advanced AI model, scientists can draw insights from music, art, and technology to analyze data from these fields to identify hidden patterns that could spark a world of innovative possibilities for material design, research, and even music or visual art.

“Graph-based generative AI achieves a far higher degree of novelty, explorative of capacity and technical detail than conventional approaches, and establishes a widely useful framework for innovation by revealing hidden connections,” says Buehler. “This study not only contributes to the field of bio-inspired materials and mechanics, but also sets the stage for a future where interdisciplinary research powered by AI and knowledge graphs may become a tool of scientific and philosophical inquiry as we look to other future work.”