The field terrain recognition based on extreme learning machine using wavelet features

摘要

Feature extraction and classification algorithm is important to determine the accuracy of classification. The terrain recognition of a legged robot has higher requirements on real-time classification. Considering the traditional training methods is difficult to meet the requirements, this paper applies the extreme learning machine using wavelet features to terrain recognition. The experimental results show that recognition rate of the extreme learning algorithm is 96.78%, which is 30.89% and 20.45% higher than BP and SVM algorithm respectively. Hence, the proposed method in this paper has obvious advantages over traditional algorithm in parameter selection and learning speed.