Improved mobile robot´s Corridor-Scene Classifier based on probabilistic Spiking Neuron Model

The ability of cognition and recognition for complex environment is very important for a real autonomous robot. A improved Corridor-Scene-Classifier based on probabilistic Spiking Neuron Model(pSNM) for mobile robot is designed. In the SNN classifier, the model pSNM is used. As network´s training, Thorpe´s learning rule is used. The experimental results show that the improved Classifier is more effective and it also has stronger robustness than the previous classifier based on Integrated-and-Fire (IAF) spiking neuron model for the structural corridor-scene. It also has better robustness than the traditional kernel-pca and the BP Corridor-Scene-classifier.