Location
USF
Document Type
Event
Keywords
inverse kinematics, gait planning, computer vision, humanoid robot
Description
Humanoid robot research and development have been an ongoing effort for many years. Tasks such as humanoid robot walking have been extensively researched, and can be often solved using control theory. In this paper, we explore how a Darwin-Op humanoid robot can autonomously balance and walk on non-flat terrains that include ramps and stairs. We use computer vision to detect the specific type of non-flat terrain based on color. Once the specific terrain has been identified, the humanoid robot computes the distance and orientation that it needs to walk before stopping. We developed the walking model using zero moment point (ZMP) trajectory planning with a cart-table model for the center of mass. We show that the robot is able to walk up the stairs and ramps, and compare experimental results with both stairs and ramps.
DOI
https://doi.org/10.5038/QDNA1747
Humanoid Robot Motion Control for Ramps and Stairs
USF
Humanoid robot research and development have been an ongoing effort for many years. Tasks such as humanoid robot walking have been extensively researched, and can be often solved using control theory. In this paper, we explore how a Darwin-Op humanoid robot can autonomously balance and walk on non-flat terrains that include ramps and stairs. We use computer vision to detect the specific type of non-flat terrain based on color. Once the specific terrain has been identified, the humanoid robot computes the distance and orientation that it needs to walk before stopping. We developed the walking model using zero moment point (ZMP) trajectory planning with a cart-table model for the center of mass. We show that the robot is able to walk up the stairs and ramps, and compare experimental results with both stairs and ramps.