Abstract With advances in technologies, autonomous systems are being used for many different applications. Nevertheless, the lack of reliability always challenges the deployment of such autonomous systems. Therefore, in parallel to efforts on increasing the degree of autonomy in new engineered systems, we should immensely improve their reliability. In this talk I will discuss a novel, systematic active-learning technique for constructing a fault diagnosis tool for a finite-state Discrete Event System (DES). The developed tool, called diagnoser, detects and identifies occurred faults by monitoring the observable behaviors of a plant. The proposed algorithm utilizes an active-learning mechanism to incrementally complete the information about the system. This is achieved by completing a series of observation tables in a systematic way, leading to the construction of the diagnoser. I then will show that the proposed algorithm terminates in a finite number of iterations and returns a correctly conjectured diagnoser. The resulting diagnoser is a deterministic finite-state automaton and is proven to consist of a minimum number of states.