Metaheuristic based navigation of a reconfigurable robot through narrow spaces with shape changing ability

Abstract

Emerging technologies and robotics solutions for cleaning applications are rapidly progressing nowadays. Reconfigurable robots that can change shapes are introduced to overcome the limitations of fixed-shape robots. Reconfigurable robots should be capable of moving through narrow spaces with the help of continuous shape-changing to improve the coverage performance. In this regard, a reconfigurable robot requires perceiving the narrow spaces in between obstacles and generating the path to navigate. Most importantly, the robot should decide and change the shape configurations appropriately to move on that path without collisions. Furthermore, such a reconfiguration ability could be used for covering around obstacles. Therefore, this paper proposes a method to navigate a reconfigurable robot through narrow spaces or cover obstacles with the aid of continuous reconfiguration. In the first step, the method analyzes the obstacle clusters in a metric map to determine the appropriate paths to navigate. In the second step, a sequence of shapes for reconfiguration that allows the robot’s movement on the path without collisions is determined. Metaheuristics, Grey Wolf Optimizer (GWO), Genetic Algorithm (GA), and Simulated Annealing (SA), are proposed to determine the shape sequence. Simulations have been conducted to validate the performance and behavior of the proposed method. The results confirm that the developed method is effective for navigating a reconfigurable robot through narrow spaces or covering obstacles with continuous shape-changing. According to the statistical outcomes, GWO requires less computational time compared to GA and SA.