A Novel Identification Methodology for the Coordinate Relationship between a 3D Vision System and a Legged Robot

• DOI: 10.3390/s150409519
• Publication Date: 2015-04-22T16:14:23Z
• Authors: Xun Chai; Feng Gao; Yang Pan; Chenkun Qi; Yilin Xu
• Keywords: Identification; Machine Vision

Coordinate identification between vision systems and robots is quite a challenging issue in the field of intelligent robotic applications, involving steps such as perceiving immediate environment, building terrain map planning locomotion automatically. It now well established that current methods have non-negligible limitations difficult feature matching, requirement external tools intervention multiple people. In this paper, we propose novel methodology to identify geometric parameters 3D mounted on without other people or additional equipment. particular, our method focuses legged which complex body structures excellent ability compared their wheeled/tracked counterparts. The can be identified only by moving relatively flat ground. Concretely, an estimation approach provided calculate ground plane. addition, relationship robot modeled. are obtained formulating problem optimization problem. integrated called “Octopus”, traverse through rough terrains with high stability after obtaining its system using proposed method. Diverse experiments different environments demonstrate accurate robust.