This paper presents localization of a mobile firefighting robot. Sensors that have been widely used for the localization in the past have shown limitations under fire environments due to low visibility and high temperatures. The extended Kalman filter was designed to accurately estimate position and orientation of the robot using relative distances to walls or objects surroundings. In addition, data from a Frequency-Modulated Continuous-Wave (FMCW) Radar, Inertial Measurement Unit (IMU) and encoders that are capable of withstanding fire environments were fused to localize the robot in indoor fire environments. For its validation, an experiment was conducted in a 2 m × 4 m area. The experimental results showed that the proposed localization method was reliable.