Comparing SONN Types for Efficient Robot Motion Planning in the Configuration Space

Motion planning in the configuration space (C-space) induces benefits, such as smooth trajectories. It becomes more complex degrees of freedom (DOF) increase. This is due to direct relation between dimensionality search and DOF. Self-organizing neural networks (SONN) with their famous candidate, Self-Organizing Map, have been proven be useful tools for C-space reduction while preserving its underlying topology. In this work, approach adapted from human motion data towards robots’ kinematics. Furthermore, three additional models are implemented compared existing ones respect abilities cover used a robot preserve total, evaluation includes six SONN architectures, representing consequent continuation previous work. Generated Trajectories reduced Output were successfully tested simulation, providing proof concept applications. The proposed method counteracts Curse Dimensionality robots many DOF, thus complete optimal algorithm can path planning.