A.I. startup Scotty Labs recently took to Medium to announce the successful completion of a $6 million seed round led by Google's A.I.-centric Gradient Ventures. According to the post, created by Co-Founder and CEO Tobenna Arodiogbu, other participants in the funding included Horizons Ventures, Hemi Ventures, Social+Capital, Neuron.VC, Gravity Ranch, and Graph Ventures. He also outlined the primary purpose of the company and some of its goals going forward. Although the company has been around for about a year, it has flown mostly under the radar in the eyes of the public. However, it may prove to be among the most important moving forward since it works on the question of how to get various autonomous driving systems up to the standards society really expects them to operate at. More directly, the company wants to accelerate progress while maintaining safety parameters that the technology certainly needs if it is going to become mainstream.
That applies to various end-products that the self-driving revolution depends on to be most efficient at solving the problems it is hoped those will solve. From self-driving cars and trucks to autonomous VTOLs and delivery robots, Scotty sees a need to advance testing methods in a way that matches the complexity of the environments those are expected to work in. Human intelligence is absolutely critical to that, the executive says. Because of that – and based on the belief that A.I., on its own, is not fully capable yet of adapting to dynamic circumstances that can arrive – Scotty Labs's solution is to start with the introduction of a platform to help the machine learning process along. It's a "tele-operations" platform that ultimately allows human operators to virtually control vehicles in dynamic environments in order to create a redundancy for A.I. to fall back on and learn from.
Scotty's first customer is already, in fact, lined up to take advantage of the platform and approach in the deployment of its own fleet – a company called Voyage. With that said, despite Scotty's drive to deliver safe self-driving cars as quickly as is safely possible, the project is still estimated to take months or years to reach level 4 autonomy. For perspective, most A.I. vehicles currently operate at level 3 autonomy and full autonomy won't be achieved until level 5 but some automakers plan to hit that point by as early as 2021, according to the latest roadmaps.