Real-Time Robotic Assistance for Elderly and Disabled Individuals at Home

The Adaptive Security and Health Analytics Response (ASHAR) system proposes a unified artificial intelligence system that can jointly tackle cybersecurity and health-monitoring issues in dynamic settings. Current hybrid systems are not very flexible and lack unified learning models, leading to inconsistent operation in changing data environments. ASHAR gets around these constraints through a federated learning-based architecture that ensures data privacy and continuously enhances model intelligence across distributed nodes. The system also uses a Natural Language Processing (NLP) companion, a Transformer-based model, to provide intuitive interaction with humans and context-aware decision support. There is also an adaptive anomaly detection engine based on bidirectional LSTM networks that can detect anomalies in security events and physiological parameters with high precision. The experiment results show that the system performs better than current systems, such as SmartGuard 2.0, AI-SecureNet, and HealthDefender, in terms of accuracy, precision, and recall. These improvements are significant (<tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{p}&lt;$</tex> 0.05) and statistically significant, indicating that ASHAR is an effective and scalable solution. The fact that it has incorporated data anonymisation and differential privacy methods is another strength, as it supports adherence to current privacy standards. Combining federated intelligence, safe learning, and humancentred communication, ASHAR provides a new paradigm that connects digital safety and individual health analytics. This study offers a flexible model that can be expanded to improve resilience within the organisational and health sector, which is likely to be used in the real-world application of the system in information-sensitive spheres.

Hali tarjima qilinmagan