Helm.ai receives $31 million to expand its “unsupervised” autonomous driving technology

Even in the face of macroeconomic challenges that have almost completely shut off the spigot of venture financing and prompted greater consolidation, there are still a few bright spots in the autonomous car sector.

One of them is Helm.ai, a startup that creates software for robots, autonomous driving, and enhanced driver assistance systems.

Just one year after securing $26 million in venture capital, the Menlo Park, California-based business recently raised $31 million in a Series C round headed by Freeman Group. Helm.ai’s valuation has increased to $431 million as a result of this most recent round, which also includes strategic partners ACVC Partners, Amplo, Honda Motor Co., Goodyear Ventures, and Sungwoo Hitech.

As part of this funding, Brandon Freeman, founder of the Freeman Group, will join the Helm.ai board of directors. To date, the company has raised $78 million.

Helm.ai was founded to advance the technology with a fresh perspective, just like so many other autonomous car firms. Tudor Achim and Vlad Voroninski, the co-founders of Helm.ai, focused on the software rather than the sensors or computing.

Similar to other companies in the sector, Helm.ai developed software that can comprehend sensor data as well as a human. The important aspect is its methodology. The so-called “brain” of the self-driving vehicle is frequently trained and improved by autonomous vehicle developers using a combination of simulation and on-road testing, as well as reams of datasets that have been annotated by humans.

According to Helm.ai, it has created software that can bypass those phases, speeding up the process and saving money. This software is especially valuable for sophisticated driver assistance systems because of the cheaper cost. The six-year-old business employs an unsupervised learning strategy to create computer programs that can train neural networks without the need for extensive fleet data, simulation, or annotation. Helm.ai is able to pitch to a wide range of customers because the software is independent of the type of computation and sensors utilized in the vehicle.

According to Voroninski, Helm.ai provides its software to a variety of OEMs and Tier 1 suppliers in the automotive sector to assist them in “achieving software differentiation with high-end ADAS and L4 solutions.”

He remarked, referring to the recent events surrounding AV sector consolidation, “Strategically, we’ve known that our go-to-market approach is going to be focused on high-end ADAS systems for some years now; our plan has not changed at all as a function of the recent events.” Since a number of years ago, I’ve basically been projecting that the vast majority of autonomous driving startups won’t succeed in reaching the market because of outmoded technological approaches and poor business models. Therefore, it has not at all surprised me. Due to all the unwarranted hoopla, the autonomous driving market hasn’t been particularly effective recently.

Numerous people have purchased Helm.ai, though Voroninski said he was unable to identify them all due to non-disclosure agreements. Honda is a client, as previously announced by Helm.ai. The mathematician who served as the previous head scientist at the cybersecurity startup Sift Security did acknowledge that he has spent the last two years concentrating on commercializing the technology and establishing alliances.

According to him, the current cash will be used for R&D, expanding those commercial agreements, and hiring more personnel to the 50-person team.