AI Made Friendly HERE

GrayMatter raises $45M for robots with ‘physics-informed AI’

Don’t miss OpenAI, Chevron, Nvidia, Kaiser Permanente, and Capital One leaders only at VentureBeat Transform 2024. Gain essential insights about GenAI and expand your network at this exclusive three day event. Learn More

Los Angeles-based GrayMatter, a startup addressing some of the hardest problems in manufacturing with AI-powered robots, today announced it has raised $45 million in a series B round of funding. The investment takes the total capital raised by the company to $70 million and has been led by Wellington Managemen with participation from multiple new and existing investors.

While robotic automation has been around for a long time, with companies like Apple using it in different functions of the assembly line, GrayMatter is pioneering what it describes as “physics-informed AI” — a technology that enables robots to self-program and handle high-mix, high-variability manufacturing environments. This is essentially the heart of the company, which has seen significant growth since its launch in 2020.

“There are so many parts, variations, and variabilities that a traditional robot cannot handle, so we’re bridging the gap with our technology for companies facing a minimum of two-year production backlogs,” Ariyan Kabir, co-founder and CEO of the company, told VentureBeat.

GrayMatter solving high-mix, high-variability manufacturing problems

The American manufacturing industry is worth $2.5 trillion, but companies are struggling with massive backlogs due to skilled worker shortages. There are as many as 3.8 million unfilled jobs across departments, keeping teams from meeting their delivery deadlines. Not to mention, in many cases, when there are enough workers, they fail to deliver the quality companies expect.

Countdown to VB Transform 2024

Join enterprise leaders in San Francisco from July 9 to 11 for our flagship AI event. Connect with peers, explore the opportunities and challenges of Generative AI, and learn how to integrate AI applications into your industry. Register Now

Kabir, who was a part of the University of Southern California’s Center for Advanced Manufacturing, saw these problems while interfacing with several industry stakeholders. The situation was even worse for companies engaged in high-mix, high-variability manufacturing dealing with a variety of parts.

This led Kabir to launch GrayMatter, with a focus on building robotic solutions that could handle labor-intensive surface treatment and finishing jobs for all sorts of products being manufactured — from football helmets to aerospace equipment and everything in between. 

At the core, the company provides enterprises with smart robotic cells, a workspace of sorts where robots using its proprietary physics-informed AI, dubbed GMR-AI, perform tasks like sanding, buffing, polishing, spraying, coating, blasting and inspection. But here is the thing: unlike automation robots that are programmed to do one specific job (which takes weeks), these machines program themselves from a high-level task description. Their process parameters adapt based on observed performance to execute the desired task autonomously. 

GrayMatter robot in action

The whole self-programming takes a matter of minutes. Once that’s done, the robots start producing highly consistent results at speed. This addresses the capacity and quality issues teams often face with manual efforts. On top of that, the cells can even monitor their health to reduce the risk of failure.

According to Kabir, GrayMatter’s physics-informed AI tries to augment existing manufacturing process models and knowledge with experimental data to deliver exactly what is expected from the robotic cell.

“It enforces known physics-based process models (or knowledge) as a constraint in the AI system to ensure that it does not learn anything that contradicts existing models/knowledge. For example, the system can enforce a constraint that increasing pressure on the sanding tool will increase the deflection of the part being sanded. We don’t need to conduct a large number of tests to learn this already-known fact. If the measured data contradicts this constraint, then it is highly likely either the sensor is malfunctioning, or the part/tool is not clamped properly,” he explained.

Adoption across different sectors

Since its launch, GrayMatter has deployed twenty custom-made smart robotic cells for enterprises in sectors such as aerospace & defense, specialty vehicles, marine & boats, metal fabrication, sports equipment and furniture & sanitary-ware.

The company did not share specific customer names, but noted these cells have cumulatively processed over 7.5 million square feet of product surface area for them.

“The work we’re doing at GrayMatter for companies…is becoming an integral part of their essential operations. It’s a big responsibility, and we’re seeing a generational shift in our lifetime. We’re in a fortunate position to be able to help millions of people elevate and improve their quality of life with our advanced AI-powered technology,” Kabir added. 

Generally, the CEO said the company’s solutions work 2-4 times faster than manual operators and cut down consumable waste by 30% or more.

In one case, an enterprise using its technology for sanding RV caps was able to bring the time taken to complete the task from one hour to six minutes per part. 

As the next step, GrayMatter plans to use this funding to scale its LA team and create next-gen AI robotic cells targeting more use cases.

“All of our current customers are asking us for adjacent products and applications because introducing our system to their production floor removes the bottleneck from that application and pushes it upstream or downstream. We have a strong product roadmap that we need to deliver. With the latest funding raise, we’re looking to create the next generation of AI robots as we continue to grow and expand our team in go-to-market, operations, product, and engineering to meet this growing customer demand,” Kabir said.

VB Daily

Stay in the know! Get the latest news in your inbox daily

Thanks for subscribing. Check out more VB newsletters here.

An error occured.

Originally Appeared Here

You May Also Like

About the Author:

Early Bird