Posts by Collection

PhD Researcher
Humanoid Robotics & Reinforcement Learning
Institute for Human and Machine Cognition (IHMC)

Duration: 2025 - Present

Perception Machine Learning Engineer
Autonomous Robots
DEKA Research and Development Corporation

Duration: January 2021 - Present

Perception Machine Learning Engineer
Medical Devices
DEKA Research and Development Corporation

Duration: January 2021 - Present

Machine Learning and Statistics Research Assistant
Nano-medicine Center (NEU)

Duration: January 2020 - May 2020

Risk Assurance Intern
Deloitte Touché Tohmatsu Limited

Duration: Summer 2018

Nuclear & Heat Transfer Equipment Systems Intern
Larsen & Toubro Limited

Duration: Summer 2017

Anticipatory and Adaptive Footstep Streaming for Teleoperated Bipedal Robots

2025 IEEE-RAS 24th International Conference on Humanoid Robots (Humanoids), 2025

Rather than replicating operator foot poses directly, we retarget user steps to robot footstep locations and predict upcoming steps to reduce timing delays — enabling seamless teleoperation synchronization on uneven terrain on the humanoid robot Nadia.

Recommended citation: Penco, L., Park, B., Fasano, S., Poddar, N., McCrory, S., Kitchel, N., Bialek, T., Anderson, D., Calvert, D., & Griffin, R. (2025). Anticipatory and Adaptive Footstep Streaming for Teleoperated Bipedal Robots. IEEE-RAS Humanoids 2025. https://arxiv.org/abs/2508.11802

Accelerating Classical Path Planning via Learned Search Space Reduction

AIAA SCITECH 2026 Forum, 2026

A learned model prunes the search space for classical path planners (A*, RRT), predicting which regions are unlikely to contain optimal paths and focusing computation where it matters — yielding significant speedups with minimal solution quality degradation.

Recommended citation: Poddar, N., Mishra, B., Clark, G., Sevil, H. E., & Griffin, R. (2026). Accelerating Classical Path Planning via Learned Search Space Reduction. AIAA SCITECH 2026 Forum. https://doi.org/10.2514/6.2026-1997 https://arc.aiaa.org/doi/abs/10.2514/6.2026-1997

Embedding Classical Balance Control Principles in Reinforcement Learning for Humanoid Recovery

IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2026), 2026

We show that injecting classical balance metrics — capture point, center-of-mass state, and centroidal momentum — as privileged critic inputs and physics-grounded reward terms enables a single RL policy to recover from any initial pose (supine, seated, kneeling, crouching) with 93.4% success, using emergent multi-contact strategies (elbows, knees, forearms) without prescribed contact schedules.

Recommended citation: Poddar, N. (2026). Embedding Classical Balance Control Principles in Reinforcement Learning for Humanoid Recovery. IROS 2026. https://arxiv.org/abs/2603.08619

Balance as Infrastructure: Environmental Contact Awareness for Loco-Manipulation Recovery in Humanoid Robots

In preparation, 2027

Extending physics-embedded RL from recovery-as-goal to balance-as-infrastructure: a single policy that walks, handles perturbations, exploits environmental contacts (walls, tables) to extend stability regions, and carries objects through recovery — without dropping them.

Automated Tetrahedron Meshing
Overcoming Numerical Challenges in Volumetric Data Processing

This paper explores the intricacies of tetrahedral mesh generation from medical volumetric data, highlighting the challenges posed by numerical errors and presenting a Delaunay triangulation-based approach for creating high-quality and subject-specific meshes.

Sinkhorn Balancing in Sudoku: A Probabilistic Approach to Constraint Satisfaction

This study explores the application of Sinkhorn balancing, a probabilistic algorithm with lower computational complexity, robustness, and ability to handle diverse problems, as an effective technique in solving Sudoku puzzles compared to traditional back propagation methods.

Analyzing the Relationship Between Crime Count, Weather Conditions, and Weekend Outings: Can They Serve as Valid Excuses for Cancellations?

This project employs statistical analysis to explore the correlation between crime count, weather conditions, and weekends, aiming to determine if these factors can be statistically significant reasons for canceling outdoor plans with friends.

YOLO v3 Object Detection Implementation and Evaluation in PyTorch

This project focuses on recreating the YOLO v3 algorithm from scratch using PyTorch, incorporating various data augmentation techniques and evaluating its performance on the MS COCO dataset, revealing challenges and insights into the model’s training and inference capabilities.

Markov Chain Modeling of Wind Speed Time Series: Analysis, Simulation, and Limitations

Explore the application of Markov chains in predicting wind speed behavior using a dataset from Chièvres, Belgium. The project includes data preparation, transition matrix computation, simulation, and autocorrelation analysis, highlighting the model’s strengths and limitations in capturing the complex dynamics of wind speed fluctuations

Modified SIR Modeling on the Spread of Ideas

This project explores the dynamics of idea sharing by introducing a system of differential equations, incorporating SIR-like and population growth models to simulate the spread of shared ideas, analyzing factors such as user flow rates, identifying potential cyclical patterns, and incorporating a capacity limit to simulate a more realistic model within a community.

Data-Driven Prediction of Aircraft Engine Remaining Useful Life for Enhanced Aviation Safety

This project focuses on leveraging machine learning techniques and the Commercial Modular Aero-Propulsion System Simulation (C-MAPSS) dataset to preprocess sensor data and develop a sophisticated model for accurately predicting the Remaining Useful Life (RUL) of aircraft engines, thereby enhancing aviation safety and operational efficiency through timely maintenance or replacement of engine components.

Patterns in Capsets

This study tackles the unsolved Capset Problem through an unconventional analysis of attribute distributions, aiming to determine the size of capsets with ‘n’ attributes and proposing that identifying specific distribution patterns could lead to a breakthrough.

Teaching experience 1

Undergraduate course, University 1, Department, 2014

This is a description of a teaching experience. You can use markdown like any other post.

Teaching experience 2

Workshop, University 1, Department, 2015

This is a description of a teaching experience. You can use markdown like any other post.