Grasshopper Optimization Algorithm: Theory, Literature Review, and Application in Hand Posture Estimation
This chapter covers the fundamental concepts of the recently proposed Grasshopper Optimization Algorithm (GOA). The inspiration, mathematical model, and the algorithm are presented in details. A brief literature review of this algorithm including differen
- PDF / 1,123,024 Bytes
- 16 Pages / 439.37 x 666.142 pts Page_size
- 108 Downloads / 227 Views
Abstract This chapter covers the fundamental concepts of the recently proposed Grasshopper Optimization Algorithm (GOA). The inspiration, mathematical model, and the algorithm are presented in details. A brief literature review of this algorithm including different variants, improvement, hybrids, and applications are given too. The performance of GOA is tested on a set of test functions including unimodal, multi-modal, and composite. The results show the ability of GOA in improving the quality of a random population, transiting from exploration to exploitation, showing high coverage of the search space, and accelerating the convergence curve over the course of iterations. The chapter also applies the GOA algorithm to a challenging problem in the field of hand posture estimation. It is observed that GOA finds an accurate configuration for a 3D hand model to match a given hand image acquired from a camera.
S. Saremi · S. Mirjalili (B) · J. Song Dong Institute of Integrated and Intelligent Systems, Griffith University, Nathan, Brisbane, QLD 4111, Australia e-mail: [email protected] S. Saremi e-mail: [email protected] J. Song Dong Department of Computer Science, School of Computing, National University of Singapore, Singapore, Singapore e-mail: [email protected] S. Mirjalili School of Electrical Engineering and Computing, University of Newcastle, Callaghan, NSW 2308, Australia e-mail: [email protected] © Springer Nature Switzerland AG 2020 S. Mirjalili et al. (eds.), Nature-Inspired Optimizers, Studies in Computational Intelligence 811, https://doi.org/10.1007/978-3-030-12127-3_7
107
108
S. Saremi et al.
1 Introduction Swarm Intelligence (SI) [1] is a field that deals with understanding and simulating the collective behaviours of organisms without a centralized control unit in nature. In this field, it is assumed that a population is made of simple agents that interact with each other and/or environment. Such agents start to interact locally and often incorporate random mechanisms to achieve a goal globally. Some of the examples in nature are: ant colony, school of fish, a flock of birds, and a herd of buffaloes. As an example, Raynold [2] simulated the swarming behaviour of birds using three simple operators: alignment, separation, and cohesion as can be seen in Fig. 1. Given the fact that each bird sees its neighbourhood only, it tries to alight its movement direction with the neighbouring birds. This means that if a predator attacks, once one bird changes its direction on the edge of a swarm, the impact cascades through the entire swarm. The key point here is that a bird in the middle or the other side of the swarm is not even aware of the predator. However, it constantly aligns its movement direction, which is good enough to avoid predators. Due to the existence of multiple neighbouring birds, the average alignment of all of them are considered. The second operator proposed by Raylond is separation which prevents collision between birds. Each bird avoids crowded
Data Loading...
