Why You’ll Thank Volvo’s E-Buses For Honking At You
We’ve all seen it happen before. The cringe worthy near-miss collisions between city buses, oblivious pedestrians and over-eager cyclists. Well, in a move to minimize the risk of accidents, Volvo has rolled out a new pedestrian and cyclist detection system for its electric buses (some of which can already be seen around European cities). This system uses a camera that constantly monitors the buses’ surroundings. Volvo says that if the system senses someone nearby, it will emit a gentle warning sound to warn the pedestrian in addition to other audio and visual cues that alert the driver as well. But, if the e-bus detects an “imminent risk of an accident” it honks very loudly, although Volvo says not as loudly as its gas-powered counterpart — something the automaker calls effective without being as disruptive.
SO, WHAT DO YOU THINK?
So, what do you think about Volvo’s pedestrian and cyclist detection systems? Could this system prove effective while being less disruptive as the manufacturer says? Tell us what you think in the comments below.
Just like humans have the ability to share their expertise with each other, robots have the potential to share skills they’ve learned with other robots instantly by transmitting the data over a network. This idea of “cloud robotics” is what Google researchers are hoping to take advantage of to help robots to gain skills even faster. Just as humans learn through trial and error, an artificial neural network follows a similar pattern. Early on, a robot’s behavior may look totally random, but by trying out different things, over time they'll learn which actions get them closer to their goals and will focus on those, continually improving their abilities. While effective, this means of programming can be time-consuming, which is where cloud robotics comes in. Rather than have every robot go through an experimentation phase, individual collective experiences can be shared, allowing one robot to teach another how to perform a simple task, like opening a door or moving an object. Periodically, the robots upload what they've learned to the server, and download the latest version, giving each robot a more comprehensive picture than any robot would have through their individual experience. By creating neural networks, teams of robots could learn and teach simultaneously, producing better results faster, opening a door to robots tackling more complex tasks sooner.
SO, WHAT DO YOU THINK?
Could this research lead to better manufacturing techniques? Tell us what you think by leaving your comments below.