The current work presents a novel approach to estimating brake surface temperature in real-time to aid in brake wear prognostics. Brake prognostics involve estimating brake pad wear in real-time, which enables its predictive maintenance. Brakes are a safety-critical system for vehicles; therefore, they require accurate and robust pad wear estimation to ensure vehicle safety. However, it involves several challenges. The estimation of pad wear is fundamentally a two-stage process: the first stage involves the accurate prediction of brake pad surface temperature, while the second stage utilizes this thermal history to calculate cumulative material wear. A significant challenge in estimating brake pad wear without an expensive sensor is that it is sensitive to the surface temperature prediction; any error in the thermal model propagates and compounds in the wear prediction stage. To identify surface temperature, traditional physical sensors are often cost-prohibitive or prone to failure in the harsh thermal and mechanical environments of the wheel end, necessitating a robust virtual sensing solution that can capture complex, non-linear heat transfer dynamics. The current work addresses the above challenge of identifying temperature dynamics using a Physics-informed Machine Learning approach. We employ Symbolic Regression (SR), a data-driven method that discovers the underlying mathematical expression of the system dynamics by searching for the optimal functional relationship between variables. SR provides an interpretable model that can be generalized across automotive platforms, offering a transparent, computationally efficient, and analytically tractable alternative to traditional ‘black box’ models. To generate the temperature dataset, a test vehicles were equipped with thermal sensors and underwent various braking scenarios. The SR-based virtual sensing model demonstrated strong and consistent predictive fidelity across all braking conditions tested. Under mild braking scenarios, the model achieved a Mean Absolute Percentage Error (MAPE) of approximately 6.0% in predicting brake surface temperature. This performance remained highly robust under mixed and harsh, high-speed braking, the most thermally demanding scenario, yielding MAPEs of only 11.6% and 11.9%, respectively.. Across all regimes, this level of temperature estimation fidelity directly limits error propagation into the downstream brake pad wear prediction stage, enabling reliable, sensor-less, cloud-based brake health monitoring at scale.