Having robots with legs who can navigate through uneven and unstructured terrain can surely provide a lot of assistance to humans. If humans exploit this capability in bipedal robots, it can be of great help in disaster management, space exploration, and even in carrying out mundane tasks on a day to day basis.
The labs at the University of California, Berkeley and Carnegie Mellon University have leveraged on the developments in nonlinear control systems as well as optimal control systems. With the help of these advances, they demonstrated ATRIAS robot who could manage walking on stepping stones, even in the case where the stones were of different heights and were randomly placed.
These robots with legs can be considered more versatile than their counterparts on wheels which have a hard time navigating through uneven terrain. There are several challenges that these robots confront when faced with difficult terrain where they need to ensure that they have a stable balance or they need to step up or down.
Quan Nguyen and Ayush Agrawal from Carnegie Mellon and UC Berkeley respectively mention why it is so tough for them. They wrote, "designing control algorithms that can handle discrete footholds (like rubble or stepping stones) is challenging, because there are strict constraints on foot placement that can't be violated and complex dynamical equations govern the motion of these systems."
They also shed light on how it was possible for robots to move on these stepping stones, “Moreover, these robots do not 'know' what the terrain will be like ahead of time; only the next step location is shown to the robot, a scenario that closely represents what a robot might encounter in the real world," they said.
The current study, “Dynamic Walking on Randomly-Varying Discrete Terrain with One-step Preview" explores "2-step periodic gait optimization.” This study aims to understand big changes in the placement of foot when a robot walks.
The team of researchers is working towards the development of frameworks to give a high degree-of-freedom to these bipedal robots. This will ensure that their footsteps are more precise on uneven terrain.
They also strive to make these robots more robust to external forces and uncertainties. To this effect, Nguyen and Agrawal mentioned in the IEEE Spectrum, “We believe that this is the first time that dynamic walking on stepping stones with simultaneous variation in step length and step height has been successfully demonstrated on a bipedal robot.”
With the experiments conducted with ATRIAS robot, the future looks very promising. The researchers are trying to combine this new development with computer vision research.
So, the robots that are currently considered “blind” can also attain a vision. Moreover, there are plans for extending these experiments to 3D walking for the robots in the real world.
These are some interesting developments in science and technology that can change the face of our earth to a great extent. If these robots are well-trained in tackling disasters or walking on the moon, it will prove to be a significant feat for humankind.