Akademik Veri Yönetim Sistemi
APPLIED SOFT COMPUTING, cilt.181, 2025 (SCI-Expanded, Scopus)
This paper introduces a comprehensive Multi-Robot Guidance Framework (MRGF) designed to simultaneously optimize task allocation and trajectory generation for aerial robot swarms operating in dynamic, obstacle-rich environments. The proposed framework introduces a novel hybrid optimization approach that combines an Injected Genetic Algorithm (IGA) with Multiple Linear Assignment (MLA) techniques, offering a scalable and efficient solution for large-scale aerial robot swarms. The MRGF aims to address the NP-hard combinatorial challenges of efficiently assigning multiple goals to a limited number of robots. By ensuring collision-free and dynamically feasible trajectories with rapid replanning capabilities, the MRGF minimizes total mission completion time and travel distance, even in complex scenarios. Extensive simulation results validate the framework's robustness and demonstrate its superiority over a couple of mature approaches in multi-objective performance metrics. Additionally, hardware-in-the-loop testing on embedded platforms confirms the MRGF's real-time capabilities even for complex scenarios with hundreds of goals and dozens of aerial robots. The MRGF achieved a mission completion time of 328 s and a total travel distance of 596 meters in a challenging simulation scenario. These findings underscore its potential for a wide range of applications, including autonomous environmental monitoring, disaster response, precision agriculture, delivery systems, and infrastructure inspection. This study highlights the MRGF as a scalable, efficient, and adaptable solution for multi-robot systems in real-world scenarios.