13 May 2016 Planning energy-efficient bipedal locomotion on patterned terrain
Author Affiliations +
Abstract
Energy-efficient bipedal walking is essential in realizing practical bipedal systems. However, current energy-efficient bipedal robots (e.g., passive-dynamics-inspired robots) are limited to walking at a single speed and step length. The objective of this work is to address this gap by developing a method of synthesizing energy-efficient bipedal locomotion on patterned terrain consisting of stepping stones using energy-efficient primitives. A model of Cornell Ranger (a passive-dynamics inspired robot) is utilized to illustrate our technique. First, an energy-optimal trajectory control problem for a single step is formulated and solved. The solution minimizes the Total Cost Of Transport (TCOT is defined as the energy used per unit weight per unit distance travelled) subject to various constraints such as actuator limits, foot scuffing, joint kinematic limits, ground reaction forces. The outcome of the optimization scheme is a table of TCOT values as a function of step length and step velocity. Next, we parameterize the terrain to identify the location of the stepping stones. Finally, the TCOT table is used in conjunction with the parameterized terrain to plan an energy-efficient stepping strategy.
© (2016) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Ali Zamani, Pranav A. Bhounsule, Ahmad Taha, "Planning energy-efficient bipedal locomotion on patterned terrain", Proc. SPIE 9837, Unmanned Systems Technology XVIII, 98370A (13 May 2016); doi: 10.1117/12.2223447; https://doi.org/10.1117/12.2223447
PROCEEDINGS
7 PAGES


SHARE
Back to Top