Heng Yang

Assistant Professor at Harvard SEAS
Robotics, Vision, Control, Optimization, Learning

hank_tall.jpg
Space Needle, 2022

I am an Assistant Professor of Electrical Engineering in the School of Engineering and Applied Sciences (SEAS) at Harvard University. I obtained my Ph.D. in Robotics from the Massachusetts Institute of Technology, where I was fortunate to work with Luca Carlone in the Laboratory for Information and Decision Systems.

I direct the Computational Robotics Lab at Harvard University. My group studies the algorithmic foundations of robot perception, action, and learning. We are broadly interested in estimation and decision-making with performance gurantees.

Current Interests: I am currently interested in three challenges.

  • Trustworthy robot perception with learned modules
  • Perception-based planning and control with performance guarantees
  • Convex optimization for estimation, decision-making, and learning

Prospective Students: For PhD opportunities, please apply to the PhD program at Harvard SEAS (e.g., in EE/CS/AM/ME) and list me as a potential advisor. I am also open to taking (undergraduate or graduate) research interns.

This year I am particularly looking for a strong candidate in statistics and machine learning.

I am interested in working with self-motivated candidates with (a) a strong theoretical and computational background (e.g., applied math, optimization and control, statistics, machine learning, and scientific computing), and/or (b) rich experiences in real robotic platforms (e.g., drones, manipulators, ground vehicles).

Due to the large volume of inquiries, I may not be able to reply to all emails, and I apologize for this.

News

Feb 6, 2024 Preprint on discounted adaptive online prediction, as a downstream application, we give a better algorithm for performing online adaptive conformal prediction!
Jan 31, 2024 Preprint on sparse polynomial optimization with unbounded sets, as an application, we solve optimal control of the Van der Pol oscillator to certifiable global optimality. Congrats to Shucheng!
Jan 8, 2024 Preprint on understanding and computing the true optimal value function of infinite-horizon pendulum swing-up
Dec 11, 2023 Talk at Department of Mathematics, National University of Singapore, thank you Kim-Chuan Toh!
Nov 30, 2023 Preprint with Yukai Tang and Jean-Bernard Lasserre: revisiting the minimum enclosing ellipsoids of set-membership estimation in control and perception, using convex optimization; my presentation at Autonomy Talks can be found here
Sep 30, 2023 Preprint with Xihang Yu: SIM-Sync, certifiably optimal 3D scene reconstruction with learned depth
Jul 30, 2023 Congratulations to Shucheng Kang on getting his paper “Verification and Synthesis of Robust Control Barrier Functions” accepted to IEEE Conference on Decision and Control!
Apr 7, 2023 Object Pose Estimation with Statistical Guarantees accepted to CVPR 2023 as a highlight paper
Mar 22, 2023 New preprint on verification and synthesis of robust control barrier functions: you do not need iterative sum of squares
Feb 23, 2023 Talk at Yau Mathematical Sciences Center, Tsinghua University

Awards & Recognitions

2022 MIT LIDS/ALL Magazine: Honing Robot Perception
2021 MIT News Spotlight: Making self-driving cars safer through keener robot perception
2021 Best Paper Award Finalist at Robotics: Science and Systems (RSS)
2021 Robotics: Science and Systems (RSS) Pioneer
2020 Graduated Non-Convexity (GNC) algorithm included in Matlab Navigation Toolbox and appeared in MathWorks News and Stories
2020 Best Paper Award Honorable Mention from IEEE Robotics and Automation Letters (RAL)
2020 Best Paper Award in Robot Vision at International Conference on Robotics and Automation (ICRA)
2019 MIT News Spotlight: Spotting objects amid clutter
2015 Tsinghua Principal Scholarship (Tsinghua News Spotlight)