This paper presents H.E.A.R.T. (Horse Equine Assisted Rehabilitation & Therapy), an innovative project that integrates Internet of Things (IoT) and Artificial Intelligence (AI) to enhance the safety and efficacy of Equine-Assisted Therapy (EAT). From an educational perspective, the project itself served as a dynamic learning framework, where engineering students applied project-based learning to develop a functional prototype, bridging theoretical knowledge with practical implementation in a real-world context. The proposed methodology centers on a smart, scaled therapy center model utilizing a NodeMCU ESP8266 microcontroller to orchestrate a network of sensors (DHT11, PIR, LDR, accelerometer, etc.) for real-time monitoring of environmental and physiological parameters. Actuators enable automated control, while an Oracle APEX cloud platform facilitates data visualization and historical analysis. A key innovative component is the integration of an AI-powered chatbot, leveraging Google's Gemini model, to provide instant, context-aware support for therapy-related inquiries. The main results confirm the system's successful performance: robust real-time data acquisition and transmission, effective automated actuator control based on predefined thresholds (e.g., fan activation at >30°C), and the development of an intuitive dashboard for remote monitoring. The project successfully demonstrates a proof-of-concept for a smarter, safer, and data-driven rehabilitation environment, while simultaneously showcasing an effective model for hands-on engineering education aligned with sustainable development goals.
