Hexapod kinematic model optimization for control system

Abstract

Control of a hexapod robot is challenging and complex task. Due to its high number of degrees of freedom (DoF) kinematic and trajectory calculations, sensor data analysis, decision making, etc. requires a lot of computing power. As most robots are becoming autonomous all calculations must be done on board and in real time. In this paper a method for hexapod robot body pose estimation is introduced. Using homogenous transformation matrices and one leg inverse kinematics an expressions are derived for robot's body rotation or shift around its body coordinate frame. Using these expressions each leg's base coordinate frame is transformed into body's coordinate frame. This method gives offset for each leg's base coordinate frame when shifting or rotating robot around its body centre. Needed leg's motor angles are calculated using one leg inverse kinematics. Also a hexapod robot control program structuring and optimization is presented. One of the key optimization aspects is body kinematic calculation optimization. All body kinematic calculations are based on 4x4 matrix multiplication which greatly loads robot's control system. These calculations can be optimized (simplified) by only using needed coordinate expressions from matrix multiplication end results.