BioSwarm: a Bio-inspired Swarm Simulator

This paper proposes an experimental simulator for bio-inspired algorithms based on Particle Swarm Optimization (PSO), and that can be applied to simulated robotics using the Robot Operating System (ROS2). The aim is to reduce the time and complexity required to evaluate swarm algorithms in scale. With this approach, the evaluation is obtained through successive trajectory optimizations to reach pre-defined targets guided by test functions. These artificial landscapes produce non-linear surfaces in the search space that allows evaluate the path planning in the simulator. The statistical results reveal that the trajectory convergence for reaching targets in the global minimum of multimodal surfaces is near-optimal.