Deep Learning Algorithm for Minimum Constraint Removal (MCR) Problem
Bo Xu, Liangwei Chen, Kewen Xu · 2020
This paper investigated the path planning problem and the feasibility of a deep learning algorithm to search a solution. In this paper, the structural model and mathematical model of the robotic path problem were constructed, and a radial basis function (RBF) neural network, which is based on the deep learning algorithm, was proposed. This algorithm extends three sets of neural networks in the direction of deep learning; the deep learning process was progressively advanced using a "reachable" network, a "least barrier" network, and a "shortest path" network. The proposed deep learning algorithm was experimentally compared with the ACO algorithm. The results indicated that the proposed algorithm can plan a path with fewer barriers and shorter distances and is suitable for solving path planning problems.