Why do we need more robots?
- by admin
In a new paper published on Monday, MIT researchers report that they have developed a method that can accurately predict the future by analyzing data about the current condition of robots and their users.
In the paper, titled “Robots and their future: Future robot behavior and user behavior”, the researchers analyze data from robots built using 3D printers to predict future behavior.
This allows for robots to be equipped with sensors and be controlled by human users, which means that robots can potentially be used to perform tasks that require greater autonomy.
The researchers’ method is based on a mathematical model that models the way robots interact with the human population.
It is an attempt to “explain how robots behave as a whole,” which is something that is “impossible to model accurately in any other way,” the researchers wrote in a press release.
“We show that our algorithm can accurately estimate the future state of a robot based on current behavior and human behavior,” said lead author Andrew Miller, a postdoctoral researcher in MIT’s Computer Science and Artificial Intelligence Laboratory.
“In other words, it can predict when a robot will behave unpredictably, and it can even predict what it will do in the future.”
The researchers found that their method, called the Neural Net, can predict the actions of robots by looking at their behavior as a function of the current state of the robot.
This is an important step because “we can then design robots with more sophisticated algorithms to improve their ability to understand the world around them and make decisions,” said Miller.
To develop their model, the researchers took data from a robot called the Piro robot, which is a popular robot used in robotics, which was built with 3D printing.
The researchers compared the robot’s behavior to other robots, which included the BeagleBee and Bebo, and their current state.
The team found that the Pico was much more accurate at predicting the future behavior of robots than other robots.
The team’s approach has been used by companies such as Google and Apple to improve the robots that they sell, but it is not yet available to the general public.
“The Piro was a great starting point, but there are many robots that are not as good as it is,” Miller said.
“With our model, we have now added an extra layer of analysis that allows us to identify better robots, and we can optimize the robots for certain functions.”
The Piro is a relatively small robot, weighing just over 200 grams, which makes it “a good target for future applications.”
While the model can be used for many tasks, the team also notes that “it cannot provide insight into how a robot’s future behavior will evolve, because we cannot know what will happen to it in the near future.”
In other words: it can’t give an indication of when the robot will go extinct or become obsolete.
“It is a wonderful achievement, and a step in the right direction, but we are still in the early days of the field,” said Andrew Miller.
“Our approach can’t predict the foreseeable future, but this paper shows that we can make predictions about the future.
I think it is important for the community to understand that we have a way to make predictions that will improve the lives of many millions of robots.”
Follow Stephanie Pappas on Twitter and Google+.
Follow us @livescience, Facebook & Google+.
Original article on Live Science.
In a new paper published on Monday, MIT researchers report that they have developed a method that can accurately predict…
- New York Times headline: ‘No one is going to stop me’: New York City mayor discusses immigration
- How to get the most out of a mobile phone app
- Walgreens’s new line of sunscreen with antibacterial ingredients is great news
- How to be a successful ‘Bachelor in Paradise’ star: The secrets to getting the perfect dress
- How to Prevent Suicide: How to Stop the Murder of Your Child