How local slopes stabilize passive bipedal locomotion?

摘要

By employing a few simple models of passive dynamic walking mechanism, we have shown the possibility of extending the boundaries of the maximum stable speed of these autonomous robots merely by changing their terrain. The replaced terrain consists of a series of parallel local slopes and is recognized as a general form of a ramp-stair surface. Although here, the mechanism of stabilization of the unstable locomotion patterns is not clearly known, the technique is quite simple and works effectively. The merit to the method over other strategies, could be described in two separate aspects: First, it is still completely passive; so we do not need any external energy to control the robot. Second, the existing passive trajectory is preserved; thus except for the robot's stability, other walking characteristics like speed, step length, period and efficiency, either do not change (if the machine is a point-foot walker) or ultimately minimally vary (if the model possesses other kinds of foot). This theory is validated using MSC Adams commercial software. There is hope that the presented passive strategy contributes to the development of efficient control algorithms, i.e. control methods boosting the gait stability without going against the gait efficiency. (C) 2016 Elsevier Ltd. All rights reserved.