Balakumar Sundaralingam

I am a Senior Research Scientist at NVIDIA. My research interests are in enabling robots to navigate and interact in unstructured environments. This includes inventing new techniques for robots, combining perception, machine learning, numerical optimization, and control theory. I also focus on efficient and practical implementation of invented techniques in modern robotic systems.

An example of my work is cuRobo, which formulates motion planning as a trajectory optimization problem and uses parallel compute on GPU to solve in 30ms. This work also leverages nvblox for collision avoidance from depth cameras. This research has been integrated into MoveIt as a planner plugin, available at Isaac ROS cuMotion.

Some of my other research efforts explored the application of sampling-based optimization for reactive manipulator control. We developed a framework (STORM) that enables optimization over non-differentiable cost terms leveraging MPPI, a sampling based optimization technique. Our framework has enabled robots to reactively avoid obstacles (CoRL 2021), move around humans to grasp objects (ICRA 2022), and also correct its behavior based on human language feedback (RSS 2022).

I received my Ph.D. in Computing (Robotics) from the University of Utah under the supervision of Prof. Tucker Hermans. My Ph.D. research focused on planning and tactile perception for in-hand manipulation leveraging trajectory optimization, learned models, and structured inference. In the past, I have worked on mobile robots, including mapping with LIDAR and reactive collision avoidance. I have also built robotic systems ranging from mobile robots to dexterous manipulation systems.

LinkedIn  /  Google Scholar  /  GitHub  /  CV

profile photo

Research Highlights

For a recent list of my research works, please visit my Google Scholar page.

project image

CuRobo: Parallelized Collision-Free Robot Motion Generation


B. Sundaralingam, S. Hari, A. Fishman, C. Garrett, K. Van Wyk, V. Blukis, A. Millane, H. Oleynikova, A. Handa, F. Ramos, N. Ratliff, D. Fox
2023 IEEE International Conference on Robotics and Automation (ICRA), 2023
arxiv / website /

project image

STORM: An Integrated Framework for Fast Joint-Space Model-Predictive Control for Reactive Manipulation


M. Bhardwaj, B. Sundaralingam, A. Mousavian, N. Ratliff, D. Fox, F. Ramos, B. Boots
Proceedings of the 5th Conference on Robot Learning, 2022
arxiv / website /

Selected for Oral Presentation (6.5% acceptance rate)

project image

Correcting Robot Plans with Natural Language Feedback


P. Sharma, B. Sundaralingam, V. Blukis, C. Paxton, T. Hermans, A. Torralba, J. Andreas, D. Fox
Proceedings of Robotics: Science and Systems, 2022
arxiv / website /

project image

Model Predictive Control for Fluid Human-to-Robot Handovers(*equal contribution)


W. Yang*, B. Sundaralingam , C. Paxton*, I. Akinola, Y. Chao, M. Cakmak, D. Fox
IEEE Intl. Conf. on Robotics and Automation, 2022
arxiv / video / website /

project image

Joint Space Control via Deep Reinforcement Learning


V. Kumar, D. Hoeller, B. Sundaralingam, J. Tremblay, S. Birchfield
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2021
arxiv / video /

project image

Robust Learning of Tactile Force Estimation through Robot Interaction


B. Sundaralingam, A. Lambert, A. Handa, B. Boots, T. Hermans, S. Birchfield, N. Ratliff, D. Fox
IEEE International Conference on Robotics and Automation (ICRA), 2019
arxiv / video / code / website /

Best Manipulation Paper Award Finalist

project image

Relaxed-Rigidity Constraints: In-Grasp Manipulation using Purely Kinematic Trajectory Optimization


B. Sundaralingam, T. Hermans
Proceedings of Robotics: Science and Systems, 2017
video / website /





Jekyll fork from Leonid Keselman's website, original design and source code from Jon Barron's website