Curious robots come in different shapes, sizes, and use a variety of sensors and cameras to dictate their movements and actions. The gadgets these robots are equipped with enable them to measure different aspects of their surroundings like depth, temperature, salinity, and several other kinds of readings. Their cameras constantly send compressed, low-resolution images to their human operators, who can redirect a curious robot to investigate anything unordinary they may come across more thoroughly.Autonomous robotic technology has only been around for 15 years, and the first software of its kind was literally built from scratch. Scientists have come a long way since then, especially with underwater bots. Most underwater robotics commonly used by oceanographers and marine biologists are only programmed to follow a specific route and monitor or search for a specific set of features. As a result, mainstream underwater robot models don’t acknowledge any anomalies they may come across, or alterations to their surroundings like temperature and terrain changes.
Described in the report as Connected and Autonomous Vehicles (CAVs), these cars have the ability to communicate with infrastructure (like traffic lights and buildings), mobile phones (even ones being used by pedestrians), and other vehicles on the road. They also have a suite of sensors that help them see and interpret the world around them, and are able to drive completely autonomously.Cities and infrastructure will have to change before autonomous cars can become mainstream. Here are four major pros and cons of these changes.
The deep-learning models are very complex. They do much more than feed map and location data to the cars. They must account for everything from terrain, to weather conditions, to road conditions, and various other environmental data, not to mention actively scan (and react to) the area and detect traffic, pedestrians, and the myriad of other things that humans encounter on the road.Intel currently has a fleet of vehicles in Chandler, Arizona, as well as autonomous driving garages – what the company calls “Labs on Wheels” — in Germany, Oregon and California.
As the Bolt slowed, then stopped for blinking yellow lights and a crosswalk at the intersection with Buchanan Street just before 9 a.m., a Toyota Highlander behind it followed suit, according to the report.A Subaru behind the Highlander failed to stop and rammed into the back of the Toyota, which lurched forward into the autonomous Bolt, according to the report.
Yet another startup tech firm has rolled into the autonomous truck arena. San Mateo, Calif.-based Embark publicly revealed its prototype self-driving truck on Feb. 24. The company, which gained approval from the State of Nevada earlier this year to begin testing its truck on public roads, said its self-driving technology enables a truck “to drive from exit to exit on the freeway without any human input.”Embark said its tractor-trailer setup uses a combination of radars, cameras and Lidar (light detection and ranging) depth sensors “to perceive the world around it.” The data points captured are processed via a form of Artificial Intelligence known as Deep Neural Nets (DNNs) that “allow the truck to learn from its own experience— much like humans learn from practice.”“Analyzing terabyte upon terabyte of real-world data, Embark’s DNNs have learned how to see through glare, fog and darkness on their own,” said Alex Rodrigues, CEO and co-founder of Embark. “We’ve programmed them with a set of rules to help safely navigate most situations, how to safely learn from the unexpected, and how to apply that experience to new situations going forward.”Embark’s truck is built specifically for “long, simple stretches of freeway driving between cities, rather than all aspects of driving.”Like other autonomous truck designs, including the Freightliner Inspiration truck and the Otto (owned by Uber) truck, Embark’s computerized truck is meant to be handed off to a human driver once it heads off the highway, who will then navigate local streets to the destination. “A human driver will still touch every load, but with Embark they’re able to move more loads per day, handing off hundreds of miles of freeway driving to their robot partners,” is how the company put it.