ReachNet : Reachability maps in the presence of obstacles
Thushara Sandakalum, Marcelo H. Ang · 2023
Calculating an accurate representation of a manipulator’s capabilities supports motion planning and control tasks. The reachability index is one of the measures used to represent a manipulator’s capabilities within the workspace but the impact of obstacles on the reachability index is not extensively explored. The aim of this manuscript was to evaluate the methods of calculating an accurate reachability index distribution amidst environment obstacles. We propose and evaluate two methods for calculating the reachability index distribution where the machine learning approach came to be the faster method which allowed the model to be applicable in real-time mobile base placement of mobile manipulators. The neural network model may also be used for calculating other distributions of workspace quality measures.