Adaptive learning and dance style simulation of sports dance robots based on artificial intelligence

Abstract The application of robotic technology in sports dance styles is becoming increasingly widespread. In response to the inability of traditional technology to adaptively learn sports dance movements, this study constructs a sports robot adaptive learning dance style simulation platform based on artificial intelligence. The random forest algorithm is used to extract sports dance movement feature, combined with the bat algorithm for training dance movements. Particle swarm optimization algorithm is used to optimize dance motion parameters and construct a dance style simulation platform. The results showed that the proposed algorithm had a mean absolute error of 0.506 and an average accuracy of 0.938 when adaptively learning dance movements. In the practical effect analysis experiment, the average score of different dance movements simulated by the proposed simulation platform reached 92.37, and the response time was only 2.249ms, significantly better than those of other dance style simulation platforms. The results indicate that this research method can achieve better dance style simulation, provide technical support for the development of dance robot technology, and demonstrate the broad potential of combining artificial intelligence and robot technology in the field of art.