Using the Bees Algorithm for wheeled mobile robot path planning in an indoor dynamic environment

• Published in: Cogent engineering Vol. 5; no. 1; p. 1426539
• Main Authors: Haj Darwish, Ahmed, Joukhadar, Abdulkader, Kashkash, Mariam
• Format: Journal Article
• Language: English
• Published: Abingdon Cogent 01.01.2018; Taylor & Francis Ltd; Taylor & Francis Group
• Subjects: Algorithms; Barriers; Collision avoidance; Collision dynamics; Computer simulation; configuration space and occupancy map; dynamic obstacle; Indoor environments; Occupancy; Optimization; Path planning; Real time; Robots; Search algorithms; static obstacle; the bees algorithm; wheeled mobile robot
• ISSN: 2331-1916; 2331-1916
• DOI: 10.1080/23311916.2018.1426539

This paper presents a solution to plan a path using a new form of the Bees Algorithm for a 2-Wheeled Differential Drive mobile robot. This robot is used in an indoor environment. The environment consists of static and dynamic obstacles which are represented by a continuous configuration space as an occupancy map-based. The proposed method is run in two respective stages. Firstly, the optimal path is obtained in the static environment using either the basic form or the new form of the Bees Algorithm. The initial population in the new form of the Bees Algorithm consists only of feasible paths. Secondly, this optimal path is updated online to avoid collision with dynamic obstacles. A modified form of the local search is used to avoid collision with dynamic obstacles and to maintain optimality of sub-paths. A set of benchmark maps were used to simulate and evaluate the proposed algorithm. The results obtained were compared with those of the other algorithms for different sets of continuous maps. This comparison shows the superiority of the new form of the Bees Algorithm in solving this type of the problems. The proposed method was also tested using AmigoBot robot. In this experiment, the proposed method was implemented using multi-threading techniques to guarantee real time performance at the dynamic stage. The results of this experiment prove the efficiency of the proposed method in a real time.