This paper deals with the real-time walking of a\r\nmulti-legged robot over difficult terrains using a balance\r\nstabilization method in order to achieve a fast speed and\r\nrobust locomotion with minimal tracking errors. The\r\nstabilization method is described through a ZMP-based\r\nonline pattern-generation scheme inspired by bio-mimetic\r\nstepping leg transferences with an active balance control so\r\nas to reduce the propagation of instability while performing\r\nrapid stepping actions for a fast walking gait in the presence\r\nof external disturbances. The proposed control system uses a\r\nforce-position controller [14] which takes impact dynamics\r\ninto consideration to compensate for the effect of external\r\nperturbations during walking by estimating impulsive\r\nforces in real-time. Using the proposed stabilization\r\nmethod, the robot plans appropriate footholds on the\r\nground in order to achieve a reasonable average walking\r\nspeed over difficult terrains in a natural environment. The\r\nsuccess and performance of the proposed method is\r\nrealized through dynamic simulations and real-world\r\nexperiments using a six-legged hexapod robot.
Loading....