Robot Path Planning Based on Hybrid Adaptive Dimensionality Representation with Glowworm Swarm Optimization

Abstract

Nowadays, robotics occupies considerable attention in human life especially in the modern industries that are automatically running. Robot path planning is considered a key problem for the robot movement. The adaptive dimensionality (AD) concept has been successfully used to solve this type of problems. Actually, algorithms of swarm intelligence optimization are largely successful in finding solutions to the robot’s path planning problems. Based on these facts, our paper aims to develop a new approach for planning the robot’s path in a hybrid state space environment which depends on AD representation and Glowworm Swarm optimization (GOS) algorithm. Also, in this paper, the effectiveness of the proposed approach will be gaged by comparing it with some other swarm intelligence algorithms. The improvement is represented by moving the robot from the start node in its environment, where the number of neighborhoods taken into consideration will be adaptive depending on the number of Neighbors of the Current node with Specific Attributes (NCwSA) and the value of the total number of neighborhoods under consideration, or the so-called Maximum Visual Range of glowworm (MVR). Accordingly, the number of neighbors from which the next node will be selected varies from one node to another. This leads to a reduction in the required storage capacity, and hence minimizing the time occupy required for processing. Adaptive dimensionality also results in decreasing the number of iterations required to determine the path from the start state to the goal state. The results prove the efficiency of the proposed approach to move the robot from start state to goal state without collision to determine the optimal path with a minimum time occupy and a minimum iterations number.