Engineering Self-Adaptive and Autonomous Systems: A Critical Survey and an Adaptation–Assurance Framework

Large Language Models for Self-Adaptive Systems: Feedback-Loop Integration and Agentic Adaptation

Adaptive self-supervised learning for real-time problem solving in autonomous systems.

The AdaptSolveAI with SelfSolveNet framework provides a practical recipe for resilient and safety-aware autonomy by enabling continuous learning from unlabeled streams, adaptation within fixed computational budgets, and action execution under control barrier function-based safety constraints.

An Automated Framework for Self-Adaptive Quality Assurance in Software Systems Using Policy-Based Reinforcement Learning

Self-adaptive information systems must keep their quality requirements while their environment changes at run time. Building the adaptation logic by hand is difficult, because design-time uncertainty makes it impossible to foresee every environmental change. Online Reinforcement Learning (RL) can build this logic automatically. However, the value-based RL methods used so far have two practical limits: the exploration rate must be tuned by hand, and continuous states must be discretised by hand. This paper presents a framework that removes both limits by using policy-based RL. The Analyze and Plan phases of the MAPE-K loop are redefined as a single policy-based decision step, and Proximal Policy Optimization (PPO) is applied for online adaptation in continuous and discrete action spaces. The framework is evaluated on two systems: a self-adaptive web application and a predictive process-monitoring system. Across four workload patterns and two concept drifts, the framework learns effective policies without exploration tuning or state discretisation. It improves on a value-based baseline (DQN) with statistical significance and performs on par with a maximum-entropy method (SAC) under the tested settings, while keeping good sample efficiency and stability.

Verify, Augment, Improve: Self-Adaptation Repair via Automated Knowledge Augmentation from Mistakes

Cyber–Physical Systems (CPSs) operate under uncertainty and cannot always guarantee the satisfaction of dependability requirements. Proactive self-adaptation mitigates violations by planning corrective actions, often leveraging predictive models. These models can be inaccurate in underrepresented regions of the operational space, leading to ineffective or unsafe adaptations. We present Verify, Augment, and Improve (VAI), a framework that extends a standard MAPE–K architecture with an asynchronous self-improving loop. VAI intercepts ineffective adaptation actions and turns explanations from a descriptive aid into a mechanism for continual improvement of the self-adaptation process. Specifically, each ineffective adaptation is verified against a ground truth (e.g., a high-fidelity simulator); when a drift between surrogate and ground truth is detected, VAI explains the failure, augments the training data near the drift, and retrains the surrogate. We instantiate VAI on a human–machine teaming benchmark and two study subjects adopting alternative ground truths. Experimental results show that VAI consistently reduces the relative error of adaptation decisions and increases the success rate of meeting requirements, with average gains of \(6.89\%\) and \(10.88\%\) across the two selected subjects.

Self-Improvements in Modern Agentic Systems: A Survey

This survey frames modern self-improving agents as adaptive systems that convert experience into accumulated capability gains, and offers a system-level framework that represents a modern agent as a configuration coupling a foundation model with an operational scaffold of prompts, memory, tools, and control logic.

Co-Evolution in Agentic Systems: Toward Self-Directed Evolution Beyond Human Design

This survey focuses on co-evolution in agentic systems, a multi-component form of self-evolution in which multiple agents and their environment impose adaptive pressure on one another.