University of Arizona, Tucson, AZ

Prasanna Thoguluva Rajendran

Ph.D. Student in Aerospace Engineering | DRL-CFD Researcher | Active Flow Control

I develop CFD-in-the-loop reinforcement learning workflows (TorchRL/Stable-Baselines3 + PyFR) for active flow control.

Projects / Research Publications CV Contact

FOCUS AREAS

Simulation, learning, and control in technically constrained environments.

DRL for Active Flow Control

Policy learning for closed-loop control in unsteady aerodynamic environments.

CFD-in-the-Loop Training

TorchRL/Stable-Baselines3 + PyFR coupling for physically grounded control policy optimization.

High-Order Discretizations

DG/FR and spectral-element methods for accurate, control-relevant simulations.

HPC Acceleration

MPI, OpenMP, CUDA, and Slurm workflows for scalable experimentation.

Scientific Visualization

ParaView/Tecplot-driven diagnostics and publication-ready technical figures.

SELECTED RESEARCH

Selected projects in DRL-CFD and active flow control.

Each entry includes problem framing, computational setup, key findings, and technical next steps.

2025 · Current Focus

Deep Reinforcement Learning for Airfoil Pitching Moment Control

Under-review Computers & Fluids manuscript on PPO-based DRL control of a NACA0012 airfoil ( $\mathrm{Re}_c=3000$ , $\alpha=10^\circ$ ) for quarter-chord pitching-moment trim using CFD-in-the-loop active flow control.

PPO
PyFR high-order FR/DG
TorchRL/Stable-Baselines3
Aerodynamic control

Read technical details

2024 · Framework

TorchRL/Stable-Baselines3 + PyFR Coupled Control Framework

Research software framework for parallelized RL-CFD experiments with reproducible training, rollout, and evaluation workflows.

Parallel environments
Automation
Experiment orchestration

Read technical details

2026 · In Progress

Optimal Pressure Sensor Placement for Reward Computation in RL-Based Active Flow Control

In-progress study on sparse pressure-probe layouts for reconstructing lift and quarter-chord pitching moment in DRL-driven airfoil flow control.

Sensor placement optimization
Sparse pressure sensing
Regression models
PyFR

Read technical details

PUBLICATIONS

Manuscripts, abstracts, and conference contributions.

Deep Reinforcement Learning for Airfoil Pitching Moment Control

P. Thoguluva Rajendran, L. Pagnier, F. Mashayek · Computers & Fluids (AI and Fluid Mechanics Symposium special issue) - DRL-AFC trim control for NACA0012 at $\mathrm{Re}_c=3000$ · 2025 · In Review

Reinforcement-Learning-Driven Active Flow Control for Airfoil Pitching Moment Trim

P. Thoguluva Rajendran, L. Pagnier, F. Mashayek · APS Division of Fluid Dynamics Annual Meeting 2025 (Session U25.00005, Low-Order Modeling and Machine Learning in Fluid Dynamics: Flow Control) · 2025 · Presented

View full publication list

Prasanna Thoguluva Rajendran

Simulation, learning, and control in technically constrained environments.

DRL for Active Flow Control

CFD-in-the-Loop Training

High-Order Discretizations

HPC Acceleration

Scientific Visualization

Selected projects in DRL-CFD and active flow control.

Deep Reinforcement Learning for Airfoil Pitching Moment Control

TorchRL/Stable-Baselines3 + PyFR Coupled Control Framework

Optimal Pressure Sensor Placement for Reward Computation in RL-Based Active Flow Control

Manuscripts, abstracts, and conference contributions.

Open to collaboration on DRL-CFD and active flow control research.