In the middle of a crisis due to Coronavirus, there are companies that keep launching great funding rounds. It is the case of the start-up Helm.ai, which has arises nothing less than $13 million in its initial round.
For years ago, the mathematician Vlad Voroninski saw the opportunity of removing some of the bottlenecks in the development of autonomous vehicle technology.
In 2016, he co-founded a start-up called Helm.ai with Tudor Archim. Currently, Helm.ai has come out of the shadows thanks to the following announcement: it has raised $13 million in its initial round with investors such as A.Capital Ventures, Amplo, Binnacle Partners, Sound Ventures, Fontinalis Partners y SV Angel.
A dozen of ‘business angels’ participated too. Among them are the founder of Berggruen Holdings, Nicolas Berggruen, the co-founders of Quora, Charlie Cheever y Adam D’Angelo, the professional NBA player Kevin Durant, the general David Petraeus and the co-founder and managing director of Matician, Navneet Dalal.
Helm.ai will allocate the $13 million of initial funding to advanced engineering and R&D and to hire more employees, as well as to close and fulfil agreements with customers.
Basically, Helm.ai is creating software that tries to understand sensor data as well as a human would, in order to drive, Voroninski said.
Such a goal does not sound much different than other companies. It is the focus of Helm.ai‘s software that is noteworthy. Developers of autonomous vehicles often rely on a combination of simulation and road testing, along with data sets that have been written down by humans, to train and enhance the so-called “brain” of the self-driving vehicle.
Helm.ai says it has developed software that can skip these steps, speeding up the timeline and reducing costs. The start-up uses an unsupervised learning approach to develop a software that can train neural networks without the need for large-scale fleet data, simulation or annotation.
“There is a very long queue and an endless sea of angular cases to go through when developing AI software for autonomous vehicles,” Voroninski explained. “What really matters is the unit of efficiency of how much it costs to solve any angle case, and how quickly it can be done. And that’s the part where we really innovate.”
“We identified some key challenges that we felt were not being addressed by traditional approaches,” Voroninski said. “We built some prototypes from the beginning that made us believe we could really take this to the end.”
Helm.ai is still a small team of about 15 people. Its commercial goal is to patent its software for two use cases – Level 2 (and a new term called Level 2+) advanced driver assistance systems found in Level 4 passenger vehicles and autonomous vehicle fleets.
Helm.ai has customers, some of whom have passed the pilot phase, Voroninski said, adding that he cannot name them.