Search
9 articles for “Proximal Policy Optimization”
-
Deep Reinforcement Learning-Based Intelligent Energy Management Strategy for Battery–Supercapacitor Hybrid Energy Storage Systems in Electric Vehicles
Abstract: As the number of EVs increases, smart solutions for energy management are needed that will optimize energy use, prolong battery life and boost vehicle performance. The application of conventional rule based and optimization-based Energy Management Strategies (EMS) for Battery–Supercapacitor Hybrid Energy Storage Systems (HESS) often leads to sub-optimal power management, supercapacitor mismatch and battery degradation when subjected to varying driving conditions. This study aims to design an intelligent energy management …
Published in International Journal of Advanced Control and System Engineering · Vol. 4, Issue 2, 2026 Read article
-
DC Motor Control using Deep Reinforcement Learning for Enhanced Robustness and Precision
Abstract: DC motors remain the workhorse of industrial automation and mobile robotics, but achieving simultaneous high-speed transient response and negligible steady-state error under variable load conditions continues to challenge classical Proportional-Integral-Derivative (PID) controllers. These model-dependent systems often require extensive tuning and struggle to maintain optimal performance when confronted with parametric uncertainties, non-linear friction, or sudden voltage fluctuations. This study presents a novel, model-free control paradigm utilizing Deep Reinforcement Learning (DRL)—specifically, a …
Published in International Journal of Electrical Machine Analysis and Design · Vol. 3, Issue 2, 2025 · pp. 22–29 Read article
-
Mario Ai Model Using Gaming Reinforcement Learning
Abstract: It is essential for research on computational and/or artificial intelligence (CI/AI) applied to games to have relevant games to apply AI algorithms to. This is pertinent. It doesn't matter if one is studying how to use CI/AI techniques to test and improve AI (e.g., games provide challenging yet scalable problems which engage many central aspects of human cognitive capacity) or how to use CI/AI techniques to improve games (e.g., player …
Published in Journal of Instrumentation Technology & Innovations · Vol. 14, Issue 2, 2024 · pp. 1–6 Read article
-
Carbon-Aware Autonomous AI Systems: Reinforcement Learning for Sustainable Cloud and Edge Computing
Abstract: The field of communication and information technology is expanding quickly. Because of this, a significant amount of carbon emissions are produced by cloud data centres and edge computing nodes. In fact they are now responsible for 3 to 4 percent of the worlds total greenhouse gas emissions. Most of the time people who manage these resources focus on how they are working and how quickly they can get things done.. …
Published in Journal of Energy, Environment & Carbon Credits · Vol. 16, Issue 2, 2026 Read article
-
Randomized Latent Vectors for Enhanced Reinforcement Learning Exploration
Abstract: This paper investigates Random Latent Exploration (RLE), a novel reinforcement learning technique that enhances exploration using randomized latent vector conditioning. I evaluate RLE’s performance across various environments, including discrete control tasks (FourRoom), continuous control (IsaacLab), and complex visual domains (Atari games). The core approach augments traditional reward functions with intrinsic rewards, calculated as the dot product between state features and periodically resampled latent vectors. The policy and value networks are …
Published in Current Trends in Signal Processing · Vol. 15, Issue 3, 2025 · pp. 19–25 Read article
-
Machine Learning Optimization for VARTM Carbon Polymer Laminates
Abstract: Vacuum-assisted resin transfer moulding (VARTM) is a key low-cost, out-of-autoclave process for manufacturing large-scale carbon-fibre reinforced polymer (CFRP) laminates crucial to aerospace wings, wind-turbine blades, marine hulls, and automotive structures. Unpredictable resin flow often leads to voids, dry spots, and race-tracking defects, resulting in 27.9% scrap rates and lengthy, costly trial-and-error design cycles. Although surrogate models provide rapid impregnation predictions for simple flat-plate geometries, vision-based monitoring is limited to idealized …
Published in Journal of Polymer & Composites · Vol. 14, Issue 3, 2026 · pp. 229–245 Read article
-
Autonomous Agentic AI for Adaptive Cure Optimization and Defect Prevention in Thermoset Polymer Composite Manufacturing
Abstract: Thermoset polymer composites occupy a central position in modern structural manufacturing, from aircraft fuselages to wind-turbine blades. Despite progress in resin chemistry and fiber architecture, the “cure process” that transforms compliant preforms into load-bearing structures remains difficult to manage. Manufacturers encounter ‘voids’, “interlaminar delaminations”, and “spring-back distortion” when curing complex or thick-section parts. The cause is not ignorance of the relevant physics, but rather that ‘temperature’, ‘chemistry’, ‘rheology’, and ‘mechanics’ …
Published in Journal of Polymer & Composites · Vol. 14, Issue 2, 2026 · pp. 301–320 Read article
-
AI-Driven Inverse Design of Functionally Graded Bio-Nanocomposites for Sustainable High-Barrier Packaging
Abstract: Multilayer plastic packaging realizes high barrier performance through laminated heterogeneous structures, but the heterogeneous structure has severe end-of-life challenges caused by the interfacial incompatibility of materials and the poor recyclability. This study proposes the inverse design of functionally graded PLA-nanoclay composite films by reinforcement learning as a monolithic alternative to traditional multilayer systems. Twin-screw extrusion is designed as a continuous control Markov decision process, and proximal policy optimization (PPO) is …
Published in Journal of Polymer & Composites · Vol. 14, Issue 2, 2026 · pp. 1547–1564 Read article
-
Neuro-Symbolic Agentic AI for Autonomous Scientific Discovery: Integrating Deep Reinforcement Learning, Quantum Simulation, and XAI-Audited LLM Hypothesis Generation in Drug Target Identification
Abstract: The exponential growth of multi-omics data and the increasing complexity of disease-associated protein interactomes have rendered conventional drug target identification pipelines computationally and epistemologically inadequate. This paper presents the Neuro-Symbolic Agentic AI for Scientific Discovery (NS-AASD) framework, a unified architecture that cohesively integrates deep reinforcement learning (DRL) exploration strategies, variational quantum simulation (VQS) of protein conformational dynamics, and XAI-audited large language model (LLM) hypothesis generation within an autonomous scientific discovery …
Published in Research and Reviews : Journal of Computational Biology · Vol. 15, Issue 2, 2026 Read article