Study On The Key Problems Of Harmful Gas Monitoring Based On Mobile Robot Platform

It is very important to monitor harmful gas, because harmful gases not only haveadverse effect on human life and the environment, but also often cause seriousaccident. This paper studied on the key problems of the harmful gas monitoringtechnique which used mobile robot as carrier. The main work includes:Firstly, the influence factor of the tunable fiber ring laser was studied, such as thelength of the erbium-doped fiber, the pump power, the intra-cavity loss and so on.Advice of appropriate value of time-delay and sampling method was given. Thispaper also proposed that the tuning range of the laser can be broadened byauto-adjusting the intra-cavity loss. Based on the study, a intra-cavity gas sensingsystem based on tunable fiber ring laser was set up, the noises of the system wasreduced, the tunable range was broadened from30nm to60nm.Secondly, data processing methods were studied in order to improve theprecision and the stability of gas detection. An adaptive Kernel method for absorptionspectrum denoising was proposed. This method did not induce distortion of thespectrum as common denoising methods do. Then a baseline fitting method wasproposed, this method only use the salient points to fit the baseline, weakened theinfluence of small absorption peaks, improved the accuracy of the baseline extraction.Thirdly, a peak detection model and a baseline fitting model were constructed, acorresponding baseline extraction algorithm was also proposed. The peak detectionmodel can detect absorption peaks of different width easily because of its multi-scalefeature. By segmented lower order polynomial fitting and segmented kernelsmoothing the fitted baseline will be smooth whilst being fidelity to the originalspectrum. Experiments show that the algorithm is effective.Last, a localization and navigation algorithm of the mobile carrier was proposed.The algorithm can locate and navigate the mobile carrier by a single landmark in asingle image, this not only minimize the sensor equipment, but also avoid themismatch of the landmark. Then, a navigation system of mobile robot platform wasconstructed. Experiments show that the algorithm was effective. A preliminaryexperiment of harmful gas monitoring was done based on the mobile robot platform.