Odor source localization research of mobile robot based on neural network

A new method for localizing odor source by mobile robot is proposed in this paper. With the information of the wind and the concentration gradient, Wasps can find odor source in a short time. However, it is very difficult for mobile robots to mimic the behaviors of wasps exactly. So, besides the bionics, BP neural network is adopted for the mobile robot to find the odor source. A simulated environment which had the information of obstacles, concentration field, the direction and the velocity of the wind and the odor source was produced by Fluent and MATLAB. In this simulated environment, the mobile robot was tested to trace the odor plumes and found the odor source. So, this algorithm is feasible and effective for the odor source localization.