This thesis compares the robustness of Maximum A Posteriori (MAP) and Infotaxis search strategies under parameter mismatch between the Bayesian update function (BUF) and actual sensor performance. The analysis was conducted using a one-dimensional (1D) single-cell beam search model. Within this framework, detection probabilities of Pᴅ ∈ {0.7, 0.9} were examined, and the estimated detection probability P̂ᴅ was varied by ±5% and ±10%. Robustness was then evaluated through 10,000 Monte Carlo simulations under these parameter settings. To validate the simulation results, experimental trials were performed using a mobile robot equipped with an ultrasonic sensor. The sensor was empirically calibrated to ensure that each measurement corresponded to a single discrete cell per iteration. The results demonstrate that both MAP and Infotaxis maintain stable performance across parameter mismatch conditions, with no significant degradation in winning frequency or average iterations under either over- or underestimation of sensor capability. Overall, the findings confirm that MAP and Infotaxis are comparable in robustness and efficiency for simple 1D search tasks. However, MAP offers a practical advantage due to its lower computational cost, making it preferable for real-time robotic implementation. The study further highlights that practical robustness for this single-cell active sensing model is dictated less by algorithmic choice and more by sensor performance.