Improving Cultural Algorithm And Its Application On Home Service Robot

Robot technology is a rapidly developing interdisciplinary research area, where the service robot is a hot research direction, as well as a breakthrough for China’s indepth study of robot technology. How to let the home service robot in complex environment be capable of family decision-making is important to free the human being while from the tedious chores.Cultural algorithm is a kind of social evolution inspired by double evolution model. Its lower population space can be combined to any intelligent optimization algorithm, while its belief space can be used to guide the population’s spatial evolution through the acquisition of information or experience from the population space, so as to improve the optimization performance of the algorithm. In this paper, the multigroup cross evolution mechanism, the mechanism of simulated annealing and adaptive parameter selection mechanism were introduced into the cultural algorithm framework, and we propose an improved cultural algorithm. This algorithm can solve global optimization problems effectively. Under the background of simulation for home server robot, we use the improved algorithm in the mission planning of home service robot. Experimental results verify the effectiveness of the algorithm. In this paper, the main work is as follows:1.We improve the cultural algorithm using the double evolutionary structure as the frame. In this fame, the multi-group cross mechanism is introduced into the the population space, and the simulated annealing mechanism is introduced into the acceptance function for extraction of outstanding knowledge in the population space to the belief space. The belief space guided the population spatial evolution by using influence function which is designed by the mechanism of adaptive parameter selection. As a results, the global search ability of the algorithm is improved through the layers of cyclic evolution. Moreover, we used the simulation of several typical function optimization results to show that the algorithm has good performance of optimization.2.According to the characteristics of service robot model, the model of service robot task planning problem is transformed into a model that can be easily solved. The improved cultural algorithm was introduced into this new model. In order to improving the efficiency and reliability of solving the problem, we added some scene analysis and task analysis strategy, and demonstrated the feasibility of this kind of problem solving method by experiment.Based on the above work, we designed the program of GDUT_TiJi Guangdong University of Technology, which better completed the task and achieved good results.