Abstract
Aiming at the problem of the search and rescue robots' perception of various gases in chemical engineering sites, a hybrid odor detection system for search and rescue robots based on particle swarm optimization was proposed. In order to improve the stability and prediction accuracy of the system, a method of using particle swarm (PSO) to optimize the weighting coefficients of the integrated neural network is proposed, that is, using the global search ability of PSO and introducing an improved PSO algorithm to optimize the weights and thresholds of the BP neural network on the basis of the original BP neural network, and the optimized network is used in the detection system, thus reducing the detection error of the system. The system analyzed the response signals of the 4 gas mixtures of the sensor array, the experimental results show that the neural network algorithm based on particle swarm optimization is applied to the training of gas mixture quantitative identification, the convergence speed is faster and the detection accuracy is higher than that of BP neural network algorithm.
