Bermuda Weed Optimization: A Scalable Meta-heuristic for Cloud-Based Cruise Control Systems

This study proposes the Bermuda Weed Optimization (BWO) algorithm; a novel and scalable metaheuristic algorithm inspired by the invasive growth of Bermuda grass. This algorithm has been developed as an enhanced version of the Invasive Weed Optimization (IWO) algorithm and, by imitating the plant's robust propagation strategies, achieves a better balance between global exploration and local exploitation. The algorithm's performance was rigorously evaluated against four previous IWO-based versions, and its superior scalability was demonstrated through the lowest average error and stable performance across diverse scenarios. Furthermore, BWO was compared with the new Gray Squirrel Search Algorithm (GSFA)—which falls outside the IWO category—to assess its performance against a novel method unrelated to the IWO family; this comparison highlighted BWO's competitive superiority and achieved an average 64.43% improvement in best-cost results. The strong convergence and scalability of BWO make it highly suitable for real-time applications—particularly in automotive systems. In a practical implementation using a cloud-based cruise control (CC) framework, BWO significantly outperformed the RPO-based method (the latest approach) by reducing overshoot by 45.92%, settling time by 29.38%, ISE speed by 8.92%, and maximum jerk by 20.09%. By achieving near-optimal convergence and leveraging cloud deployment with high scalability, BWO can effectively adapt to diverse automotive system requirements and achieve high efficiency across multiple operating modes.