Adaptive robotics marks a fundamental shift in artificial intelligence and intelligent control, positioning robots not just as programmed agents but as autonomous systems capable of learning and evolving within dynamic and uncertain environments. This evolution is pivotal not only for control robustness but also for enabling autonomous agents to exhibit real-time learning and adaptive reasoning. This new generation of robots leverages the integration of artificial intelligence, machine learning, metaheuristic algorithms, and fuzzy logic, enabling continuous interaction among perception, decision-making, and action.

At its core, adaptive robotics relies on a data-driven approach that used real-time sensor data to identify parameters, refine internal models, and adapt control strategies thereby minimizing reliance on rigid analytical formulations. Key tools in this process include artificial neural networks for modeling nonlinear dynamics and enabling predictive behaviors, fuzzy inference systems for managing ambiguity and environmental noise, and metaheuristic algorithms for structural optimization and fine-tuning control policies in high-dimensional search spaces.

This approach is vital in domains such as autonomous robotics, human-robot interaction, and mobile systems, where variability and unpredictability are common. Adaptive robotics allows machines to operate effectively in scenarios where prior models are incomplete or rapidly changing, making it a fundamental technology for next-generation intelligent automation.

This research topic aims to address the limitations of static, model-based robotic systems by advancing methodologies and applications in adaptive robotics, underpinned by computational intelligence. Traditional robotic control architectures often fail when confronted with high environmental variability, unforeseen perturbations, or the need for real-time adaptation. The increasing demands in fields such as intelligent manufacturing, service robotics, and field robotics require systems that are not only autonomous but also self-improving. Adaptive robotics, as explored in this topic, emerges as a powerful alternative—one that leverages learning algorithms, data-driven modeling, and evolutionary computation to create robots that can self-calibrate, generalize from experience, and make intelligent decisions with limited prior knowledge. Building on recent advances in data-efficient learning, neuromorphic systems, and the real-time integration of AI, this topic invites contributions that explore both theoretical and applied aspects of adaptive robotics. The focus lies in how computational intelligence methods can drive intelligent behavior, increase robustness, and promote generalization across diverse operational contexts.

Relevant areas of interest for this research topic include, but are not limited to, the following:

• AI-based Adaptive Control: Methods combining machine learning, fuzzy logic, and parameter estimation for robust real-time adaptation in robotic systems.

• Metaheuristic Optimization in Robotics: Algorithms for controller tuning, behavior evolution, and structure optimization under uncertainty.

• Data-Driven Modeling and System Identification: Approaches for identifying robot dynamics and environmental models from real-time sensor data.

• Artificial Neural Networks in Control and Perception: Development of ANNbased architectures for decision-making, motion planning, and prediction.

• Fuzzy and Hybrid Logic Systems: Integration of fuzzy inference systems with neural or evolutionary models to enhance interpretability and adaptability of robotic systems.

• Applications in Mobile Robotics, Human-Robot Interaction, and Autonomous Systems: Case studies demonstrating adaptive behavior in navigation, manipulation, or multi-agent coordination.

• Interdisciplinary and Cross-Domain Approaches: Research connecting adaptive robotics with neuroscience, cognitive systems, or human-centered design

Are you interested in contributing a manuscript to this Research Topic?

Adaptive robotics: A computational intelligence perspective