A research group from Tohoku University in Japan has, for the first time, successfully demonstrated that by changing only its speed parameters, a quadruped robot will spontaneously change its steps between energy-efficient patterns, exhibiting the phenomenon known as gait transition.
Until now, the manner in which changing speeds cause quadrupeds to change their gaits—from walking to trotting to galloping—was poorly understood. Through this study, the researchers successfully demonstrated the reproduction of the quadruped gait transition. They achieved this via a decentralized control scheme, using a simple local rule in which a leg continues to support the body while sensing weight on the other legs.
(A) The quadruped robot, Oscillex 3. (B) Reproduced changes in step (gait transition phenomenon). (Top) Change in the parameter related to speed (Bottom) The colors represent periods when a leg senses weight on the other legs. The gait transition was demonstrated from walk to trot, to canter, to gallop. (Image courtesy of Akio Ishiguro.)
Moreover, they confirmed that the energy-efficiency profile of the robot's gait patterns matched those measured in horses.
The present work is expected to lead to a better understanding of the mechanism of how quadrupeds can flexibly and efficiently adjust their gait when their speed is changed. The results may constitute the basis of an unconventional approach to coordinating the many degrees of freedoms required for adaptive robot locomotion.
The researchers hope that their study will lead to a wide range of applications such as adaptive legged robots working in disaster areas, user-friendly legged entertainment robots, and automatic motion-creation algorithms for computer graphics (CG) animation.
Centralized control approach: A main computer governs all degrees of freedom (DoFs) on the body at all times to track the desired trajectory of each point, resulting in high computational cost and low adaptability to environmental changes. Decentralized control approach: Each local controller for a leg modulates its locomotor rhythm with the use of only local load sensing, resulting in low computational cost and high adaptability. (Image courtesy of Akio Ishiguro.)
Their research is published in the journal Scientific Reports.
For more quadrupedal robotics news, check out this video of Boston Dynamics’ newest bot.
Source: Tohoku University