Sangli Teng

I'm a postdoctoral researcher at the University of California, Berkeley, advised by Prof. Koushil Sreenath. I received my Ph.D. in Robotics from the University of Michigan in May 2025, advised by Prof. Maani Ghaffari and Prof. Ram Vasudevan. I received my B.Eng. in Mechatronics from Chongqing University in 2020. I was a visiting researcher at MIT SPARK Lab led by Prof. Luca Carlone in 2024.

Email  /  Scholar  /  X (Twitter)  

Research

I design geometric algorithms with certifiable performance guarantees for autonomous systems.

Honors & Awards

• [07/2026] Best Paper Award 🏆 RSS 2026 "The Geometry of Motion" Workshop
• [05/2026] ICML 2026 Gold Reviewer
• [05/2026] Best Paper Award Finalist 🏆 IEEE RAS TC on Optimization for Robotics
• [03/2026] ProQuest Distinguished Dissertation Award Honorable Mention 🏆
• [06/2025] RSS 2025 Outstanding Reviewer Award 🏆
• [06/2025] Most Popular Poster Award 🏆 RSS 2025 "Equivariant Systems" Workshop
• [04/2024] Rackham Predoctoral Fellowship
• [07/2023] Best Paper Award Finalist 🏆 RSS 2023
• [08/2020] University of Michigan Robotics Institute Fellowship

Selected Publications

	LieIPM: Lie Group Interior Point Method for Direct Trajectory Optimization of Rigid Bodies Sangli Teng, Ruiqi Zhang, Tzu-Yuan Lin, William A Clark, Mark Mueller, Ram Vasudevan, Maani Ghaffari*, Koushil Sreenath* In submission for future publications A fast IPM solver dedicated for rigid bodies leveraging the Lie group structures. * Equal advising
	Embedding Hybrid Systems into Continuous Latent Vector Fields Sangli Teng, Hang Liu, Koushil Sreenath International Conference on Machine Learning (ICML) 2026 Best Paper Award, RSS 2026 workshop: The Geometry of Motion: Physics-Informed Structures for Learning and Control Hybrid systems can be described by continuous ODEs.
	CHyLL: Learning Continuous Neural Representations of Hybrid Systems Sangli Teng, Hang Liu, Jingyu Song, Koushil Sreenath Transactions on Machine Learning Research (TMLR) The flow of hybrid system admit a higher-dimensional continuous representation.
	Ego-Vision World Model for Humanoid Contact Planning Hang Liu, Yuman Gao, Sangli Teng, Yufeng Chi, Yakun Sophia Shao, Zhongyu Li, Maani Ghaffari, Koushil Sreenath IEEE International Conference on Robotics and Automation (ICRA), 2026 A vision-conditioned world model for humanoid contact-rich motion planning.
	Learning Hybrid Dynamics Via Convex Optimizations Kaito Iwasaki, Sangli Teng, Anthony Bloch, Maani Ghaffari American Conrol Conference (ACC), 2026 Convex optimization-based identification of hybrid dynamical systems.
	Max Entropy Moment Kalman Filter for Polynomial Systems With Arbitrary Noise Sangli Teng, Harry Zhang, David Jin, Ashkan Jasour, Ram Vasudevan, Maani Ghaffari, Luca Carlone Advances in Neural Information Processing Systems (NeurIPS) 2025 Most Popular Poster Award, RSS 2025 Equivariant Systems Workshop Best Paper Award 2025 Finalist, IEEE RAS TC on Optimization for Robotics Optimal Kalman-type filter for nonlinear non-Gaussian systems.
	Invariant Filtering for Full-State Estimation of Ground Robots in Non-Inertial Environments Zijian He, Sangli Teng, Tzu-Yuan Lin, Maani Ghaffari, Yan Gu IEEE/ASME Transactions on Mechatronics, 2025 State estimation for legged robot on moving platforms.
	Riemannian Direct Trajectory Optimization of Rigid Bodies on Matrix Lie Groups Sangli Teng, Tzu-Yuan Lin, William A. Clark, Ram Vasudevan, Maani Ghaffari Robotics: Science and Systems (RSS), 2025 Constrained singularity-free on-manifold optimization for rigid bodies.
	Discrete-Time Hybrid Automata Learning: Legged Locomotion Meets Skateboarding Hang Liu, Sangli Teng, Ben Liu, Wei Zhang, Maani Ghaffari Robotics: Science and Systems (RSS), 2025 Learning hybrid systems as the inductive bias for interpretible reinforcement learning.
	A Generalized Metriplectic System Via Free Energy and System Identification Via Bilevel Convex Optimization Sangli Teng, Kaito Iwasaki, William Clark, Xihang Yu, Anthony Bloch, Ram Vasudevan, Maani Ghaffari American Conrol Conference (ACC), 2025 Bilevel system identification of a symmetric structure for Hamiltonian systems.
	Convex Geometric Motion Planning of Multi-Body Systems on Lie Groups Via Variational Integrators and Sparse Moment Relaxation Sangli Teng, Ashkan Jasour, Ram Vasudevan, Maani Ghaffari International Journal of Robotics Research (IJRR) RSS 2023 Best Paper Award Finalist Plan full rigid body motion with verifiable global optimality and feasbility.
	Convex Geometric Trajectory Tracking Using Lie Algebraic MPC for Autonomous Marine Vehicles Junwoo Jang, Sangli Teng, Maani Ghaffari IEEE Robotics and Automation Letters (RA-L), 2023 A QP-based MPC for surface vehicle trajectory tracking.
	Fully Proprioceptive Slip-Velocity-Aware State Estimation for Mobile Robots Via Invariant Kalman Filtering and Disturbance Observer Xihang Yu, Sangli Teng, Theodor Chakhachiro, Wenzhe Tong, Tingjun Li, Tzu-Yuan Lin, Sarah Koehler, Manuel Ahumada, Jeffrey M. Walls, Maani Ghaffari IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2023 Fuse wheel kinematics and disturbance model for slip detection of ground vehicle.
	Input Influence Matrix Design for MIMO Discrete-Time Ultra-Local Model Sangli Teng, Amit K. Sanyal, Ram Vasudevan, Anthony Bloch, Maani Ghaffari American Control Conference (ACC), 2022 Model-free tracking control of fully-actuated systems.
	Lie Algebraic Cost Function Design for Control on Lie Groups Sangli Teng, William Clark, Anthony Bloch, Ram Vasudevan, Maani Ghaffari IEEE Conference on Decision and Control (CDC), 2022 Geodesic-based symmetry-preserving cost function design on matrix Lie groups.
	An Error-State Model Predictive Control on Connected Matrix Lie Groups for Legged Robot Control Sangli Teng, Dianhao Chen, William Clark, Maani Ghaffari IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2022 Error-state MPC for legged robot trajectory tracking.
	Toward Safety-Aware Informative Motion Planning for Legged Robots Sangli Teng, Yukai Gong, Jessy W. Grizzle, Maani Ghaffari Combining safety awareness and information gain in motion planning for legged robots.
	Legged Robot State Estimation in Slippery Environments Using Invariant Extended Kalman Filter With Velocity Update Sangli Teng, Mark Wilfried Mueller, Koushil Sreenath IEEE International Conference on Robotics and Automation (ICRA), 2021 Fuse vision and kinematics for legged robot state estimation.

Template adapted from Jon Barron's website.