Considering radial basis function neural network for effective solution generation in metaheuristic algorithms

In many engineering optimization problems, the number of function evaluations is severely limited by the time or cost constraints. These limitations present a significant challenge in the field of global optimization, because existing metaheuristic methods typically require a substantial number of function evaluations to find optimal solutions. This paper presents a new metaheuristic optimization algorithm that considers the information obtained by a radial basis function neural network (RBFNN) in terms of the objective function for guiding the search process. Initially, the algorithm uses the maximum design approach to strategically distribute a set of solutions across the entire search space. It then enters a cycle in which the RBFNN models the objective function values from the current solutions. The algorithm identifies and uses key neurons in the hidden layer that correspond to the highest objective function values to generate new solutions. The centroids and standard deviations of these neurons guide the sampling process, which continues until the desired number of solutions is reached. By focusing on the areas of the search space that yield high objective function values, the algorithm avoids exhaustive solution evaluations and significantly reduces the number of function evaluations. The effectiveness of the method is demonstrated through a comparison with popular metaheuristic algorithms across several test functions, where it consistently outperforms existing techniques, delivers higher-quality solutions, and improves convergence rates.

Performance Evaluation of Bundle Adjustment with Population Based Optimization Algorithms Applied to Panoramic Image Stitching

The image stitching process is based on the alignment and composition of multiple images that represent parts of a 3D scene. The automatic construction of panoramas from multiple digital images is a technique of great importance, finding applications in different areas such as remote sensing and inspection and maintenance in many work environments. In traditional automatic image stitching, image alignment is generally performed by the Levenberg–Marquardt numerical-based method. Although these traditional approaches only present minor flaws in the final reconstruction, the final result is not appropriate for industrial grade applications. To improve the final stitching quality, this work uses a RGBD robot capable of precise image positing. To optimize the final adjustment, this paper proposes the use of bio-inspired algorithms such as Bat Algorithm, Grey Wolf Optimizer, Arithmetic Optimization Algorithm, Salp Swarm Algorithm and Particle Swarm Optimization in order verify the efficiency and competitiveness of metaheuristics against the classical Levenberg–Marquardt method. The obtained results showed that metaheuristcs have found better solutions than the traditional approach.

Comparative Solutions of Exact and Approximate Methods for Traveling Salesman Problem

There are two major optimization methods: Exact and Approximate methods. A well known exact method, Branch and Bound algorithm (B&B) and approximate methods, Elimination-based Fruit Fly Optimization Algorithm (EFOA) and Artificial Atom Algorithm (A3) are used for solving the Traveling Salesman Problem (TSP). For 56 destinations, the results of total distance, processing time, and the deviation between exact and approximate method will be compared where the distance between two destinations is a Euclidean distance and this study shows that the distance of B&B is 270 , EFOA is 270 and A3 is 288.38 which deviates 6.81%. For time processing aspect, B&B needs 12.5 days to process, EFOA needs 36.59 seconds, A3 needs 35.34 seconds. But for 29 destinations, exact method is more powerful than approximate method.