In this study, experimental attempts were made to reduce exhaust gas toxic emissions from the spark ignition (SI) engine. For this, a sucrose catalyst is coated inside the metallic substrate. The obtained emission level was compared with the results of commercial catalysts for lean-burn operations. The engine was operated at 20%, 40%, 60%, 80%, and 100% loads and the highest NOx conversion efficiency of 60.217% at 40% engine load and 70.732% of hydrocarbon (HC) conversion efficiency at 100% engine load was achieved. Exhaust emissions from the sucrose-coated catalytic converter are observed to be lower than the conventional commercial converter. Also, this paper attempts to predict the emission characteristics of both rigs kept under observation using a fuzzy logic expert system (FLES). Both the input and output responses from the real-time SI engine are used to train and test the proposed FLES. The FLES proposed in this study can predict the emission characteristics of both conventional and sucrose coated catalytic converters with an accuracy of 97%.