Robust and efficient forward, differential, and inverse kinematics using dual quaternions

Modern approaches for robot kinematics employ the product of exponentials formulation, represented using homogeneous transformation matrices. Quaternions over dual numbers are an established alternative representation; however, their use presents certain challenges: the dual quaternion exponential and logarithm contain a zero-angle singularity, and many common operations are less efficient using dual quaternions than with matrices. We present a new derivation of the dual quaternion exponential and logarithm that removes the singularity, we show an implicit representation of dual quaternions offers analytical and empirical efficiency advantages compared with both matrices and explicit dual quaternions, and we derive efficient dual quaternion forms of differential and inverse position kinematics. Analytically, implicit dual quaternions are more compact and require fewer arithmetic instructions for common operations, including chaining and exponentials. Empirically, we demonstrate a 30-40% speedup on forward kinematics and a 300-500% speedup on inverse position kinematics. This work relates dual quaternions with modern exponential coordinates and demonstrates that dual quaternions are a robust and efficient representation for robot kinematics.

Automated summary

This study introduces a new derivation of the dual quaternion exponential and logarithm that removes its zero-angle singularity, demonstrates that the implicit representation of dual quaternions offers higher efficiency advantages compared to matrices and explicit dual quaternions, and derives dual quaternion forms of differential and inverse position kinematics. Analytically, implicit dual quaternions are more compact and require fewer arithmetic instructions for common operations. Empirically, the speedup ranges from 30-40% for forward kinematics to 300-500% for inverse position kinematics.