Authors: Gaikwad Mayuri Santosh, Dr.Anand Khatri, Professor Bangar Abhishek Popat, Professor S.B.Bhosale
Abstract: Ayurvedic Dravya identification is an important process in Ayurveda, as the correct identification of medicinal substances supports their appropriate use and quality assessment. However, identifying Ayurvedic Dravya based on sensory characteristics such as Rasa (taste) can be challenging because different substances may exhibit similar taste profiles, while their responses can vary with sample concentration, solvent composition and preparation conditions. Traditional sensory evaluation also depends on human perception and may introduce subjectivity. An electronic tongue (E-tongue), which uses an array of chemical sensors to generate measurable response patterns, offers a potential approach for objective and repeatable taste-related analysis. This paper proposes a Reference-Anchored Multi-Dilution Electronic Tongue framework for Ayurvedic Dravya identification. The proposed system evaluates each sample at multiple predefined dilution levels to obtain a more representative sensor-response fingerprint. The collected signals are preprocessed, normalised and compared with reference fingerprints stored in a database using a deterministic distancebased matching method. Unlike approaches that depend on artificial intelligence or machine-learning classification, the proposed framework focuses on reference-based comparison, controlled sample preparation, sensor calibration and predefined acceptance criteria. The system is designed to provide the closest reference match when the similarity criteria are satisfied and a “No Reliable Match” result when the evidence is insufficient. The paper also discusses key challenges, including sensor drift, variation in sample composition, solvent effects, dilution consistency and the need for laboratory reference validation. The proposed framework may support the development of a transparent and reproducible electronic-tongue system for Ayurvedic Dravya screening. Experimental validation with authenticated samples is required to establish its identification performance, repeatability and practical suitability.
