Estimation of a Plume with an Unmanned Terrestrial Vehicle

This work involves the design and implementation of a gas-sensing mobile robot as an experimental tool to localize a carbon dioxide source. The autonomous robot achieves navigation through an embedded microcontroller using a strap-down accelerometer and a fusion of four CO2 sensors. A mass flow controller and diffuser are used to dependably generate a plume that simulates a point source. A base station receives sensor data and calculates the robots position using the accelerometer data filtered using a low pass filter followed an Extended Kalman Filter. This method has applicability for unmanned vehicles tracking emissions of contaminants and their effects in the environment