Nandan Tumu

I am a 5th-year PhD student at the University of Pennsylvania, passionate about advancing the frontiers of robotics through the integration of human knowledge and intuition with machine learning. My research spans three key areas: uncertainty quantification using Conformal Prediction, physics-informed machine learning, and the application of these techniques to multi-agent robotics and game environments. By integrating these approaches, I aim to develop learning methodologies for robotics which are more trustworthy, and have lower sample complexity. I am fortunate to be supported by the NSF Graduate Research Fellowship.

news

Jul 2024	Our paper on conformal off-policy prediction was accepted to CDC 2024!
Mar 2024	Our paper on designing practical non-conformity scores for planning and control was accepted to L4DC 2024!
Mar 2023	Our paper on Physics-constrained learning for trajectory prediction with Conformal Preduction was accepted to Intelligent Vehicles 2023!
Jan 2022	We won the Japan Automotive AI Challenge, against teams from Honda, Nissan, Toyota, and more!

latest posts

Jun 22, 2024	Multi-Modal Conformal Prediction Regions with Simple Structure by Optimizing Convex Shape Templates
Jun 06, 2022	Autoencoder-based Perspective Transformation and Segmentation for Autonomous Racing
Jul 10, 2019	Feature Scaling in Real Cardiac Electrograms

selected publications

Multi-Modal Conformal Prediction Regions by Optimizing Convex Shape Templates

Renukanandan Tumu, Matthew Cleaveland, Rahul Mangharam, and 2 more authors

Dec 2023

Abs HTML

Conformal prediction is a statistical tool for producing prediction regions for machine learning models that are valid with high probability. A key component of conformal prediction algorithms is a non-conformity score function that quantifies how different a model’s prediction is from the unknown ground truth value. Essentially, these functions determine the shape and the size of the conformal prediction regions. However, little work has gone into finding non-conformity score functions that produce prediction regions that are multi-modal and practical, i.e., that can efficiently be used in engineering applications. We propose a method that optimizes parameterized shape template functions over calibration data, which results in non-conformity score functions that produce prediction regions with minimum volume. Our approach results in prediction regions that are multi-modal, so they can properly capture residuals of distributions that have multiple modes, and practical, so each region is convex and can be easily incorporated into downstream tasks, such as a motion planner using conformal prediction regions. Our method applies to general supervised learning tasks, while we illustrate its use in time-series prediction. We provide a toolbox and present illustrative case studies of F16 fighter jets and autonomous vehicles, showing an up to \68}% reduction in prediction region area.
Physics Constrained Motion Prediction with Uncertainty Quantification

Renukanandan Tumu, Lars Lindemann, Truong Nghiem, and 1 more author

In 2023 IEEE Intelligent Vehicles Symposium (IV) , Jun 2023

Abs HTML

Predicting the motion of dynamic agents is a critical task for guaranteeing the safety of autonomous systems. A particular challenge is that motion prediction algorithms should obey dynamics constraints and quantify prediction uncertainty as a measure of confidence. We present a physics-constrained approach for motion prediction which uses a surrogate dynamical model to ensure that predicted trajectories are dynamically feasible. We propose a two-step integration consisting of intent and trajectory prediction subject to dynamics constraints. We also construct prediction regions that quantify uncertainty and are tailored for autonomous driving by using conformal prediction, a popular statistical tool. Physics Constrained Motion Prediction achieves a 41% better ADE, 56% better FDE, and 19% better IoU over a baseline in experiments using an autonomous racing dataset.