Robotic arm may allow wheelchair users to open doors

From Softpedia.com:

Experts in the field of robotics have had a hard time making their own inventions open doors, on account of the difficult tasks that are involved in the process, but now a team has managed to crack the problem for a real-life application. While it may seem simple for us, a great degree of coordination and intelligence is needed to open a door. Former University of Massachusetts in Lowell (UML) student Erin Rapacki has managed to come up with a low-cost solution that allows for the mounting of a specialized gripper on wheelchairs, which makes it possible for users to navigate through doors without having to reach them themselves.

When opening a door, our brains take into account a number of complex factors, and put them in relation, in order for us to be able to open the “obstacle” in the first attempt. We need to know our exact position as opposed to the door, and also the extent of our own hand. We then must calculate “how much force is needed to open the door, the twisting angles to unlatch the door, and how much force is needed to unlatch it,” Rapacki says. This holds true for robots as well, only, in their case, humans need to devise the algorithms that allow them to perform such intricate calculations fast enough to make them of use.

Another important function that robotic grippers need to perform is being able to catch on a wide variety of door-knob shapes and sizes. As we look at the knobs, we almost instantly understand how they work, and what needs to be done to make them open. But a robot is having a much tougher time trying to do just that, specialists say. The current attempt at creating such a system is not the first, obviously, but general-purpose, wheelchair-mounted robotic arms come in very high price ranges, and the UML team wanted to create a low-cost solution. Its innovation was presented last week, at the IEEE robotics conference, NewScientist reports.

DORA (door-opening robotic arm) proved remarkably efficient in handling the tests that were laid down in front of it. It managed to unlatch doors using about 14 different types of handles. It was also successful in 85 percent of the instances where it had to push the door to open it, and in 65 percent of the cases where it had to pull on it, which is admittedly more difficult. But the best part about DORA is the fact that it was only built under a $2,000 price tag, a massive improvement (cost-wise) from general-purpose robotic arms. Granted, it is specialized on a single function.