A Review of Control Strategies for Robotic Systems with Embedded Sensors

Abstract

Robotic systems increasingly rely on embedded sensors to achieve high levels of autonomy, adaptability, and safety in dynamic environments. Unlike existing surveys, this review provides a comprehensive and structured taxonomy of control strategies for sensor-integrated robots, presenting a novel comparative perspective that links traditional methods with emerging paradigms. Five key categories are systematically analyzed: classical controllers, model-based techniques, sensor-driven feedback schemes, data-driven learning approaches, and bio-inspired intelligent control. A detailed comparative evaluation highlights their respective advantages, limitations, and suitability for different robotic domains. Beyond summarizing established techniques, this work contributes a forward-looking analysis of recent technological advances, including sensor miniaturization, edge AI for onboard learning, self-calibrating and self-tuning controllers, explainable control frameworks, and IoT-enabled cyber-physical integration. Furthermore, this review identifies standardized benchmark datasets, evaluation metrics, and simulation platforms to support reproducibility and rigorous performance assessment. By consolidating fragmented literature and emphasizing emerging trends, this study uncovers critical research challenges, such as real-time multimodal sensor fusion, robust and machine learning-based controllers, and energy-aware architectures for embedded robotics. This review establishes a comprehensive reference framework for researchers and practitioners, offering actionable insights that can accelerate the development of resilient, interpretable, and high-performance robotic control systems.