Comparison of Conventional and Modified Ant Colony Approaches for Path Planning of Robot in an Indoor Environment

Abstract

The optimal path for autonomous robots is highly needed, particularly in the industrial communities, manufacturing and social support industries. The meta-heuristic methods are strong and adaptable, and their employment in autonomous problems is taken into consideration. One of the methods used by the researchers is Ant Colony Optimization (ACO). However, this approach is less effective due to a few limitations. Thus, in order to mitigate its shortcomings and enhance the robustness of path planning, this paper aims to compare the different standard ACO algorithm with the new modified built algorithm, which is developed by taking the concept of ACO and extending it by condition-based rules, named as condition based-ant colony concept (CB-ACC), to solve path planning. The theoretical foundation of ACO with modification as another approach are used in this study to illustrate the ACO methods in the path planning, particularly in the static-complex environment. Simulation results show the comparison of the ACO approaches, including the Ant System (AS), the Ant Colony System (ACS) and the modified condition based-ant colony concept (CB-ACC), and how they are effective for static-complex environments compared to conventional ACO algorithms. Moreover, this paper presents the comparison results of the three approaches.