Driving Sustainability Through Machine Learning – Mobility Moments With Secondmind CEO Gary Brotman

Founded in 2016 by a team of mathematicians and engineers, the UK’s Secondmind uses machine learning to helps automotive innovators achieve greater sustainability. In 2020, the Cambridge-based company entered into a strategic partnership agreement with Mazda to explore how it can accelerate time to market for Mazda’s fuel-efficient cars

On this week’s Mobility Moments, we talk to Secondmind’s CEO, Gary Brotman, about how AI can be utilised to design cleaner cars.

How did you get into the AI and mobility space?

I’ve spent the last 20+ years commercialising, and evangelising product and service lines for leading artificial intelligence, mobile, internet and technology companies including Secondmind, Yahoo! and Qualcomm, where I was the Head of AI Strategy and Product Planning and led the development and commercialization of the company’s AI Engine across the Snapdragon Mobile Platform portfolio.

My areas of expertise range from public relations to product management within two broad technology domains, digital music and machine learning.

I was first introduced to machine learning as the enabling technology powering music recommendations more than 20 years ago and the impact it had then motivated me to learn more and to seek out opportunities to leverage AI and machine learning in the design and development of new products and services.

Describe Secondmind’s key products and services?

Secondmind exists to help innovators in automotive design better cars in less time and achieve sustainability through machine learning.

We make this possible today with the Secondmind optimization platform, which uses state-of-the-art machine learning capable of achieving significant improvements in powertrain design and calibration in the face of the mounting complexity created by increasing emissions standards, performance and consumer demands, and sustainability targets.

How can AI (and Secondmind) help achieve greater sustainability?

Secondmind’s state-of-the-art machine learning helps engineers break through the barriers of complexity and fundamentally transform the model-based development process. In Powertrain optimization, and more specifically, engine ECU calibration, early indications are that our unique approach can help car makers reduce data acquisition and processing costs by up to 80% and reduce calibration time by up to 50% by automating the design of experiments.

The precision of our advanced machine learning modelling, combined with our unique Active Learning approach is capable of significantly reducing prototype material costs by up to 40%.

As a result, the potential impact to both business and environmental sustainability is substantial.

Describe Secondmind Labs

Secondmind Labs is led by our Chief Science Officer and Chairman, Carl Edward Rasmussen, Professor of Machine Learning at Cambridge University. Under his leadership, our team of researchers and engineers uses proven mathematical principles to build scalable tools that solve problems for our customers.

We have combined branches of maths and ML engineering in ways that have never been done before, and then applied them to solve some of the most challenging problems facing automotive engineers today.

What are your thoughts on the conclusion of the COP26 summit?

Car makers remain under intense pressure to produce vehicles based on a broad array of complex powertrain configurations designed to meet consumer and bottom-line demands.

These pressures are compounded by increasingly more stringent and necessary emissions regulations. Achieving business and environmental sustainability with such headwinds will require radical new approaches to transforming model-based development to get ahead of the curve.

A primary driver of climate tech investment and the key to accelerating this evolution is more rapid adoption and deployment of practical, state-of-the-art machine learning throughout the design and development process. This new breed of software adeptly manages increasing complexity. It enables faster, automated experiments that compress critical, time-consuming and costly steps in the development process from months to weeks.

Dramatic efficiencies like these will unlock more investment for electrification and other strategic growth areas to ensure long-term sustainability for car markers and the planet.