Efficient Environment Exploration for Multi Agents : A Novel Framework

Abstract

View Video Presentation: https://doi.org/10.2514/6.2023-1088.vid This study introduces Adaptive Aquila Optimization Algorithm, designed particularly for Multi-Agent Space Exploration. By constructing a finite map, this enables the acquisition of a collision-free optimum mobility path in a barrier-filled environment. The proposed framework based upon Adaptive Aquila Optimizer (AAO) is a unique blend of deterministic Coordinated Multi-agent Exploration (CME) with a modified swarm-based technique namely Aquila optimizer (AO). The conventional Aquila optimization algorithm is dynamically modified to improve the rate of exploration by introducing stochastic parameters. The architecture, also known as the Coordinated Multi-robot Exploration Adaptive Aquila Optimizer (CME-AAO), begins by utilizing deterministic CME to determine the cost and utility values of neighbouring cells surrounding the agents. The rate of exploration is then further enhanced using adaptive Aquila optimization. The efficiency of the proposed CME-AAO algorithm was subsequently validated by performing wide range of simulations under various environmental conditions. The obtained output is then checked with contemporary algorithm CME-Aquila based algorithm. Results indicate that the proposed CME-AAO framework significantly improves map navigation in a congested environment in a comparatively shorter exploration time. This makes the proposed methodology ideal for use on-board in an dynamic environment, where conventional optimizer methods either fails or takes considerably longer duration to converge

Topics & Concepts