In this work, we adapted the Whale Optimization Algorithm (WOA) to a mobile robot system to map the chemical gas concentration in an unknown environment. It is assumed that robots do not know the environment beforehand but can measure it with the existing sensors on them. The chemical gas concentration in the environment is modeled by a multimodal Gaussian function (a function composed of a sum of multiple Gaussian functions). In addition, we assume that the sensor measurements of the robots are corrupted by a normally distributed random noise and investigate the effect of the noise on the performance of the algorithm. While WOA acts as a high-level path planning algorithm for the agents, low-level navigation and collision avoidance of the robots with unicycle dynamics is achived utilizing potential functions. The measurements of the robots are filtered and interpolated to obtain the chemical concentration map of the environment. Successful performance of the algorithm is verified through simulations performed in the MATLAB environment.