Odor Plume Direction Estimation via Gas Sensor Array

The direction of the odor plumes is very helpful for robots to preform odor source localization task. To address issues of that wind measurement equipment is not able or convenient to carry on the robots, this paper proposes a method for estimating the incoming direction of the odor plume with a gas sensor array. Firstly, a model for estimating the incoming direction of the odor plume was established via the positional relationship between gas sensors in the sensor array. Secondly, by using the output responses of the gas sensors in the array, the time delay between gas sensors was calculated via cross-correlation method. Thirdly, the solution of the model was solved using the Levenberg-Marquette method. Finally, the results of gas sensor arrays in different distribution patterns indicate that our proposed method can estimate the incoming direction of odor plume effectively.