Home automation systems are evolving rapidly with advancements in machine learning (ML) and gesture-based controls, enhancing user convenience and safety in smart environments. Traditional automation interfaces lack intuitive control and adaptability, making integrating ML-driven gesture recognition and safety monitoring essential. Eventbased sensors capture dynamic, high-resolution data, allowing asynchronous gesture interpretation and optimized device control. ML models, including convolutional and recurrent neural networks, improve gesture recognition, while safety monitoring systems identify hazards like falls and dangerous proximity for vulnerable residents. This paper comprehensively reviews current advancements in gesture control, safety monitoring, and energy efficiency within home automation. It highlights the efficacy of technologies such as event-based cameras, sensor networks, and ML algorithms while addressing limitations in accuracy under varying environmental conditions. A comparative analysis suggests future improvements in adaptive, secure, and energy-efficient systems that support personalized, user-centered automation.