This paper presents a fresh swarm-based metaphorical approach termed the Yellow Banded Wasp Algorithm (YBWA) that mimics the adaptive communal dynamics of yellow-banded wasps. The YBWA employs the competition-driven territorial search (CDTS) and the memory-based updating procedure, allowing a balance between exploration and exploitation. The CDTS enables the wasps to adopt a decentralized approach to decision-making, while the memory archive retains the elite solution for future iterations. When utilized together, these aspects simplify the population diversity and the convergence dynamics to the global optimum. The adeptness of YBWA has been meticulously benchmarked on the IEEE-CEC-2017 test functions across dimensions 10 and 30, and the IEEE-CEC-2022 test suite over dimensions 10 and 20. The suggested YBWA outperforms eleven present-day optimization approaches (OAs) in terms of the mean fitness value across all the function types. The YBWA also highlights more rapid and stable convergence, with a smaller standard deviation throughout the repeated runs. A chain of non-parametric assessments has been carried out to substantiate the exclusivity of the YBWA. Furthermore, the YBWA has been employed to assess the six rigid transformation components for the Point Cloud Registration (PCR) utilizing the Bildstein Station 1 dataset. The YBWA yields an exact spatial alignment of the point clouds, characterised by low registration error and an optimized set of transformation components. Hence, the YBWA is a viable and reliable approach to be deployed as an optimizer for the registration process and optimistic tasks (OTs).
Agrawal et al. (2026) studied this question.