After spearheading the development of autonomous self-driving cars for several years, Google LLC is now turning its attention to commercial shipping.

The internet giant announced last week that it’s going to work with Rolls-Royce Holdings plc to help improve the engineering firm’s intelligent awareness systems for ships. The partnership will see Rolls-Royce use Google’s Cloud Machine Learning Engine to train its artificial intelligence-based object classification system that’s used to detect, identify and track things that ocean-going vessels are likely to encounter.

In other words, Rolls-Royce is using Google’s platform to create technology that will almost certainly be used to power autonomous ships. Rolls-Royce has made no secret of its desire to built a fleet of autonomous vessels, but current technologies such as radar and the Automatic Identification System aren’t reliable enough to facilitate them because they don’t always spot everything a vessel might encounter at sea.

For example, small boats such as kayaks don’t show up on radar screens, and neither would a kayak be able to transmit AIS data to tell the ship it’s nearby. At present, spotting these kinds of things is the job of the ship’s crew members, who are on board keeping watch at all times.

Google’s Cloud Machine Learning Engine uses the same neural net machine intelligence technology that powers many of its own services, including image and voice search. The idea is that if this technology is combined with cameras placed all around a ship, it might become more efficient than human lookouts are. One day, the technology could even remove the need for humans to keep a lookout.

“While intelligent awareness systems will help facilitate an autonomous future, they can benefit maritime businesses right now making vessels and their crews safer and more efficient,” Rolls-Royce ship intelligence Senior Vice President Karno Tenovuo said in a statement. “By working with Google Cloud, we can make these systems better faster, thus saving lives.”

Besides the cloud machine learning deal, Google and Rolls-Royce will also conduct a joint research project on multimodal and unsupervised machine learning. They also said they’ll carry out research to determine if speech recognition and synthesis systems can be used in human-machine interfaces for marine applications. The companies will also collaborate to optimize the performance of local neural network computing on ships using software such as TensorFlow, which is Google’s open-source machine intelligence software library.

According to Rolls-Royce, the development of AI-based awareness systems can only make vessels “safer, easier and more efficient to operate,” as they’ll provide crews with a better understanding of their vessel’s surroundings.

Image: Rolls-Royce