An Efficient and Improved Particle Swarm Optimization Algorithm for Swarm Robots System
In recent years, the number of researches in which swarm intelligence shown by individual communication in swarm robots is increasing. As one of the representative algorithms in swarm intelligence, particle swarm optimization has been applied to many fiel
- PDF / 1,798,974 Bytes
- 9 Pages / 439.37 x 666.142 pts Page_size
- 103 Downloads / 236 Views
Abstract In recent years, the number of researches in which swarm intelligence shown by individual communication in swarm robots is increasing. As one of the representative algorithms in swarm intelligence, particle swarm optimization has been applied to many fields because of its simple concept, easy realizing and good optimization characteristics. However, it still has some disadvantages such as easy falling in the local best situation and solving the discrete optimization problems poor. In this paper, genetic algorithm has been integrated with particle swarm optimization to improve the performance of the algorithm; the simple particle swarm optimization algorithm has been simulated in the Player/Stage and compared with the particle swarm optimization. The simulation shows that the algorithm is faster and more efficient. Keywords Swarm robots
Particle swarm optimization Simulation analysis
1 Introduction Swarm intelligence is a method to achieve artificial intelligence by imitating biological group behavior in the natural world [1], which offers a new thought to the solutions of complex issues by using group advantage without centralized control and global model [2]. Z. Shi (&) X. Zhang J. Tu Z. Yang School of Computer and Communication Engineering, University of Science and Technology Beijing, Beijing, People’s Republic of China e-mail: [email protected] X. Zhang e-mail: [email protected] J. Tu e-mail: [email protected]
Z. Yin et al. (eds.), Proceedings of The Eighth International Conference on Bio-Inspired Computing: Theories and Applications (BIC-TA), 2013, Advances in Intelligent Systems and Computing 212, DOI: 10.1007/978-3-642-37502-6_40, Springer-Verlag Berlin Heidelberg 2013
329
330
Z. Shi et al.
The researches of swarm robot systems drew lessons from optimization techniques and principles such as swarm intelligence. It is an application of swarm intelligent in the multi-robot system [3], which is also a new research field in which multi-robot system is given the nature of group in the usual sense [4]. As a representative of the swarm intelligent algorithms, particle swarm algorithm and swarm robots searching are the instances of the smart agent searching. And there is a certain mapping relation between them. Therefore, swarm robots in the real world can be modeled and simulated by the use of particle swarm optimization [5]. In order to improve the convergent speed and accuracy of the algorithm, Genetic Algorithm (GA) has been integrated with PSO in this paper. The simple particle swarm optimization (sPSO) Algorithm which proposed by Hu Wang et al. combined with obstacle avoidance principle has been simulated in the Player/Stage and compared with the bPSO in the timeline.
2 Research Background and Related Work Genetic algorithm, one of the three major branches of evolutionary computation, is a kind of adaptive, probabilistic, random and iterative search algorithm. It is evolved through the reference from the evolution law of the biosphere. Particle swarm algorithm is used to solve con
Data Loading...
