ABSTRACT
The goal of kinematic identification is to estimate a kinematic parameter vector ρ that accounts in the least-squares sense for positioning and orientation errors of the robot. A common approach to the problem has been to define such a set of variables in terms of additive changes dρ to the robot nominal link parameter vector ρ 0 in a given kinematic model. Linearization of the robot forward kinematic equations about ρ 0, at the particular ith joint configuration qi , provides an Identification Jacobian matrix J i = J ( q i , ρ 0 ) https://s3-euw1-ap-pe-df-pch-content-public-u.s3.eu-west-1.amazonaws.com/9781315138725/3769e96e-c54b-409c-981a-8dd0c38c57b7/content/eq223.tif"/> . The Jacobian relates yi , the vector of end-effector pose errors at the ith configuration to dρ, the vector of independent kinematic parameter errors, () y i = J i d p https://s3-euw1-ap-pe-df-pch-content-public-u.s3.eu-west-1.amazonaws.com/9781315138725/3769e96e-c54b-409c-981a-8dd0c38c57b7/content/eq224.tif"/>
