The effort shows how low-cost drones and robotic systems--combined with rapid advances in machine learning--are making it possible to automate whole sectors of low-skill work. Avitas uses drones, wheeled robots, and autonomous underwater vehicles to collect images required for inspection from oil refineries, gas pipelines, coolant towers, and other equipment. Nvidia's system employs deep learning, an approach that involves training a very large simulated neural network to recognize patterns in data, and which has proven especially good for image processing. It is possible, for example, to train a deep neural network to automatically identify faults in a power line by feeding in thousands of previous examples.
Engineers at the University of California, Riverside (UCR) have developed a new online energy management system (EMS) that they say can improve PHEV fuel efficiency by more than 30%. In "Development and Evaluation of an Evolutionary Algorithm-Based Online Energy Management System for Plug-In Hybrid Electric Vehicles," published in IEEE Transactions on Intelligent Transportation Systems, Xuewei Qi and colleagues explain that improving the efficiency of current PHEVs is limited by shortfalls in their energy management systems (EMS), which control the power split between engine and battery. The EMS developed by Qi and his team combines vehicle connectivity information (such as cell networks and crowdsourcing platforms) and evolutionary algorithms – a mathematical way to describe natural phenomena such as evolution, insect swarming and bird flocking. "We combined this approach with connected vehicle technology to achieve energy savings of more than 30 percent.
You can add one more name to the constantly expanding list of companies that want a slice of that autonomous driving pie, as a new company named Drive.ai The new company, which also announced that it has added former General Motors Vice Chairman and Board Member Steve Girsky to its Board of Directors, is looking to put its stamp on the self-driving space with its own deep learning algorithms. These full stack deep learning algorithms, Drive.ai CEO Sameep Tandon says that the team at Drive.ai has been working on these deep learning applications since the company was founded in 2015. Its focus at the start will be on outfitting route-based industries with its technology, which includes the self-driving system itself, along with a collection of sensors, an interface for the driver of the vehicle, and roof-mounted communication system. The team's roots are based in Stanford University's Artificial Intelligence Lab, so it certainly sounds like Drive.ai has the know-how to put together a system such as this and make its name known in the world of autonomous driving.