The new chip will teach the robots to anticipate the movement

To stand in the way of an industrial robot was a stupid idea. These cars are moving quickly and accurately, but they are blind and stupid, so happy to break your hand and strikes the skull, if you find yourself in their way. So the testers had to have the courage to try this trick with the robotic arm in Realtime Robotics, a startup located in the heart of the Boston seaport. The tester leaned forward to intercept the movement of the robot, when he grabbed the item off the table and sent it into the box.

Fortunately, the robot paused, deftly circled her outstretched arm and gently cupped the object in the box. Broken hand today.

Such graceful agility can be incredibly useful for robotics. Some robots can work together with people, but they are usually low-powered, inaccurate, and used sparingly. The most powerful and capable industrial machines still have to work in very precisely controlled conditions, away from the soft, brittle people.

Why industrial robots are dangerous to humans?

Some companies are trying to find ways to work around this problem. Some test sensors that will stop the powerful robots in their way if they detect an obstacle. Realtime Robotics goes further, giving the robots the intellectual ability of a low level required to move on to the real world. This physical awareness of people and animals take for granted every time when you raise your arm or leg.

In several rooms Realtime Robotics industrial robot hands have capability of the new chip developed by the company. When connected to three-dimensional sensors such chip allows the machine to quickly consider a number of different actions, in fact “imagining” the result before you select the most suitable for the task.

“The main problem is that robots are very stupid,” says George Konidaris, founder and chief robotics Realtime Robotics. “We have this basic motor skills, as robots don’t have”.

Planning of movements is deceptively tricky for a robot, partly because each joint adds an extra dimension to the calculations that must be performed.

The chip company performs mathematical calculations using relatively simple scheduling algorithm of the movement, was developed by Konidaris and his colleagues, when he still studied at Duke University. During the parallel execution of computations allocated to the chip can perform more than 10 000 faster than a standard computer chip, while consuming less energy.

It is part of a broader trend. Advances in hardware and software will gradually lead to higher IQ robots, thereby opening up new possibilities for use. More intelligent robots can sit on the production line next to a man — knowing how to pick up items regardless of their location and without accidentally causing harm to anyone. It can speed up automation in many industries.

Improved motion planning will be fundamentally important for the future of robotics.

“Any robot that intentionally moves to achieve the objectives should have a better understanding of how he should move,” say the researchers. “The problem is that the planning of movements is slower, if the environment is complicated, especially when the robot has many degrees of freedom”.

There is another great potential application of this technology: self-driving cars. Just as the robot needs to plan its movement, if it wants to avoid collision, and self-driving car needs to choose the safest route around obstacles. Realtime Robotics is already developing for this special chip. Konidaris says that this should allow cars with independent management to quickly adapt to challenges on the road may make them safer.

It all depends on decide whether the manufacturers of manipulators to use the technology for Realtime Robotics. Some are already testing it. The ultimate success will depend on integration.

Not far off the hour when robots will no longer interfere with people and will be safe for them. Agree? Tell us in our chat in Telegram.

Leave a Reply

Your email address will not be published. Required fields are marked *