AI is democratising aerodynamics: tests that once required a wind tunnel can now be carried out virtually

Netcompany INEOS will incorporate a new aerodynamic analysis tool in 2027 capable of creating digital replicas of its riders and evaluating different positions and equipment through simulations. The technology, developed by the American company AiRO, promises to multiply the possibilities of conducting tests without constantly relying on a wind tunnel.

AI opens a new era in cycling aerodynamics without the need for wind tunnels

Aerodynamics has become one of the major fields of development in professional cycling. Teams dedicate resources to find small improvements in their riders' positions, helmets, or equipment, but conducting those tests requires time, specialized facilities, and a considerable investment.

Artificial intelligence is opening new possibilities in this area. According to the company AiRO, confirmed to Cyclingnews, Netcompany INEOS will use its technology during the 2027 season. The agreement coincides with the presentation of Vayu, a new aerodynamic simulation tool that expands the possibilities of its platform.

RECOMENDADO

Alan Hatherly extends his Jayco AlUla deal until 2028 and remains determined to return to the top of MTB

Automatic brake light, rear radar and 20 hours of battery life: meet the new Lezyne Radar Pro 300

Red Bull-BORA-hansgrohe backs cyclocross and opens the door to a multidisciplinary project – including MTB?

We test the new Specialized Ambush 3: more ventilation, a better fit and an adjustable peak once again

Lucinda Brand signs for Canyon//SRAM and picks Paris-Roubaix 2028 for her road cycling farewell

Seixas casts doubt on his participation in the 2027 Tour de France and does not rule out targeting the Giro d’Italia

Although the collaboration is confirmed by the technology provider, the specific details of its application within the British team have not yet been announced. AiRO expects to provide more information in the coming days.

How AiRO works and why it doesn't need a wind tunnel to compare positions

The AiRO technology is based on the idea of creating a three-dimensional representation of the cyclist and using it to study how air flows around them.

The process begins with photographs of the rider and some physical data, such as their height and weight. From that information, an artificial intelligence system generates what the company calls a Digital Aero Twin, a digital replica that allows reproducing their morphology and modifying their posture.

Then, computational fluid dynamics, known as CFD, comes into play. This technology uses numerical calculations to simulate the behavior of air and estimate the aerodynamic drag generated by the cyclist.

According to the technical documentation of AiRO, creating the three-dimensional model takes approximately three minutes, while each aerodynamic simulation is completed in about seven minutes using high-performance servers.

The platform allows comparing different trunk, arm, or head positions and evaluating the influence of certain helmets. The results help estimate changes in CdA, the coefficient of aerodynamic drag multiplied by the frontal area, and its possible translation into watts saved at a given speed.

An important peculiarity is that AiRO's usual simulations do not include the complete bicycle. The company itself explains that it primarily calculates the resistance of the cyclist and uses an additional value to represent the bicycle. This simplification reduces the time and cost of calculations, although it also limits the accuracy of absolute values.

Vayu will allow exploring hundreds of combinations of helmets and positions

The new tool Vayu aims to significantly expand the number of configurations that can be analyzed.

During a presentation to several media outlets, Ingmar Jungnickel, founder of AiRO and former head of aerodynamics at Specialized, explained that one of the possibilities in development is to compare more than 20 helmets on the same cyclist, using different angles and head positions.

This would allow studying more than 400 virtual combinations without having to physically change helmets and repeat each test in a wind tunnel.

However, this massive comparison function is still under development, so it should not be confused with a fully available and validated capability.

Another of the possibilities presented consists of analyzing more than a hundred small posture modifications, including variations in arm, shoulder, or head positions, to identify those that could reduce aerodynamic drag.

The goal is not only to find a theoretically faster position. Afterwards, the specialist must check that the rider can reproduce and maintain it while pedaling without compromising their ability to generate power or their comfort.

INEOS could conduct aerodynamic tests with its entire roster

One of the main advantages of this system for a WorldTour team is the possibility of conducting a much larger number of tests.

Traditionally, wind tunnel sessions focus on certain riders or moments of the season. The available time and costs make it difficult to repeat them constantly with the entire roster.

With a digital replica of each cyclist, INEOS could evaluate new positions or configurations without needing to move all its riders to specialized facilities.

Jungnickel hopes that the team will use AiRO as a complement to traditional tests, especially to explore alternatives and select those that are worth physically verifying later.

It also opens the door to studying position modifications that require an adaptation process. A configuration may be aerodynamically excellent but too demanding to maintain for several hours or cause discomfort. In that case, trainers and biomechanics specialists would have to assess its viability.

Is it as accurate as a wind tunnel?

The technology does not start from scratch. AiRO already offers its services through specialists in biomechanics and cycling positioning, with sessions costing around 200 dollars, according to Cyclingnews, although the final price depends on the center.

The company has also published a validation report in which it compares its simulations with measurements taken in a wind tunnel.

In a test of 30 positions corresponding to different sports disciplines, it achieved a very high correlation between both systems, with an R² of 0.9987. However, it also detected a systematic difference of 24% between the calculated and measured absolute drag values.

This is relevant because it indicates that the system can be particularly useful for identifying which position offers the least resistance, but not necessarily for accurately reproducing the absolute CdA value obtained in a wind tunnel. Moreover, that validation included disciplines other than cycling and was conducted without a bicycle.

The company itself acknowledges other limitations of its technology, such as the absence of dynamic pedaling in the usual simulations, the use of headwind, and the difficulty in accurately reproducing certain fabrics, clothing wrinkles, or the internal ventilation of helmets.

Therefore, the arrival of Vayu does not mean that the wind tunnel has become obsolete.

What is interesting about the agreement with INEOS is that this technology can change the way aerodynamic tests are organized. Instead of using specialized facilities to explore each modification, the team could first study dozens of alternatives on the computer and reserve physical tests to verify the most promising solutions.

If it manages to translate those advantages into the daily work of its riders, INEOS will have a new tool to seek aerodynamic gains. It remains to be seen which specific functions of Vayu will be operational in 2027 and to what extent the British team will incorporate these simulations into its regular development program.