The issue of allowing robots to walk is something engineers and researchers have been working on for some time, and now researchers from Switzerland research institute École Polytechnique Fédérale de Lausanne (EPFL) are applying novel algorithms to the problem.
While walking is a natural ability that humans have, the ability to do so on two legs is actually a complicated task. Several muscles must perform delicate balancing acts. Robots tend to either step cautiously or fall over easily.
In the quest for more natural bipedal robotic walking, the team from EPFL's Biorobotics Laboratory are working on a Compliant Humanoid (COMAN) that use new walking algorithms focused on improving the robot's balance while it is walking. The control algorithm is based on the existing symmetries in the structure and dynamics of the robot, as well as the mathematical equations representing the robot dynamics.
According to Hamed Razavi, a researcher scientist at the Biorobotics Lab, the robot effectively has a lot more information about its own structure. The control algorithm uses sophisticated computer programs to carefully analyse the date received from the robot – such as its position, velocity, and joint angles – and sends appropriate commands to the motors, telling them what to do in order to maintain the robot's balance.
“You could say we’re working in harmony with these symmetries rather than against them. As a result, we obtain a more natural and robust walking gait,” says Razavi.
If COMAN is pushed, for example, the algorithms will calculate where its foot should land in order to stop itself from falling.
COMAN also has special joints, which are integrated with elastic elements, providing it with greater flexibility when performing different tasks.
The intention is for robots equipped with such algorithms to carry out rescue missions. According to the researchers, a bipedal robot would be able to get around a lot easier than wheeled robots, especially in environments designed by humans for human, with elements such as doors and stairs.
Another application would be to have the robots carry heavy boxes or move objects. And finally, the robots could be converted into exoskeletons for the disabled.
The next step is refining the algorithms so the humanoids have a wider range of movement and can overcome obstacles and walk on irregular or sloped surfaces.