Scientists Create Robot that Swaggers Like a Human

Scientists at the Georgia Institute of Technology have created a robot named DURUS that can walk a lot like a human, and with the swagger of one with a lot of confidence.
Although it may seem like a task that scientists would have mastered years ago, it is actually very difficult to create a fully bipedal robot, especially one that is not only battery powered, but also fully mimics human footsteps. In the past, bipedal robots have been created, which often walk flatfooted and hunched over. DURUS is the first to have a more upright posture and walk heel to toe, which is a lot more like the natural gait of a human. [1]
Aaron Ames, associate professor of automation and mechatronics says that the process is a lot more complicated than one would expect. Getting the robot to have a natural human swagger requires some intense work.

“In terms of equations, these things are insane. If you were to write them down, we would have thousands of pages of equations that underline this basic motion of the robot,” Ames said. [2]

However, research into the creation of robots that appear like humans has shown that bipedalism is actually fairly ineffective, despite the fact that we humans rely on it. Robots are much more likely to stay steady and not fall over if they have multiple legs or operate on wheels.
So while a multi-legged robot might function better than that of a two-legged robot, the robots themselves are designed to mimic humans in order to respond in disaster scenarios at home and on the battlefield. They must walk on two legs in order to squeeze into tight spaces to perform a rescue. Also, having more than two legs or operating on wheels can make it impossible for a robot to navigate stairs, which might render it useless in many scenarios.
One of the biggest pitfalls of the robot at the moment is that once it falls, it is unable to get back up again. This would make it extremely difficult to use it in a genuine disaster scenario. However, scientists are currently working on remedying this problem.
Sources:
1. Gizmodo
2. Washington Post