New website: https://hankyang94.github.io/
Before that, I will be a Research Scientist in the Autonomous Vehicles Research Group at NVIDIA led by Marco Pavone starting July 2022.
I'll build a ComputationalRobotics lab studying the algorithmic foundations of robot perception, action and learning.
Contact me if you are interested in pursuing a PhD or Postdoc. I am particularly interested in candidates with
- a strong theoretical and computational background (e.g., applied math, optimization, statistics, machine learning, and numerical computing), AND/OR
- rich experiences in working with real robotic platforms such as drones, manipulators, and ground vehicles.
However, feel free to reach out if you are passionate about solving problems related to Computational Robotics, and I'd love to hear your thoughts.
I will also be slowly migrating my website and providing more information. Stay tuned!
My name is Heng (Hank) Yang, and I am currently a PhD candidate at MIT Laboratory for Information and Decision Systems (LIDS) and Department of Mechanical Engineering (MechE). I am excited to be working with Prof. Luca Carlone in the the SPARK (Sensing, Perception, Autonomy, and Robot Kinetics) Lab on foundamental robotic perception, computer vision, and optimization and learning algorithms.
I am broadly interested in robotics, computer vision, optimization, and machine learning. My long-term research vision is to enable safe and trustworthy autonomy for a broad range of high-integrity applications (e.g., autonomous driving, space robotics), by designing tractable and provably correct algorithms that enjoy rigorous performance guarantees, developing fast implementations, and validating them on real robotic systems.
My PhD research focused on designing what we call certifiable algorithms for outlier-robust geometric estimation in robot visual perception. Despite the NP-hardness of the mathematical optimization problems involved in outlier-robust geometric estimation, certifiable algorithms are polynomial-time algorithms that offer provable optimality guarantees. I have established the theoretical and computational foundations of certifiable perception based on a wide range of advanced tools (e.g., robust estimation, semidefinite relaxation, large-scale convex optimization), and successfully demonstrated the trustworthiness of certifiable algorithms on safety-critical applications such as self-driving and space robotics.
My future research aims to start from certifiable geometric perception, and reach my vision of system-level safe and trustworthy autonomy, by building an advanced toolbox combining theory, computation, and system validation. I plan to execute two important steps towards this vision. The first step aims to integrate certifiable perception with deep feature learning to achieve safe perception, and the second step aims to integrate safe perception with safe action to construct system-level safety guarantees. You can refer to my research statement for more details.
Before joining the SPARK Lab, I worked with Dr. Brian Anthony in the Device Realization Lab on designing a low-cost and portable ultrasound shear wave elastography device using external mechanical vibration (see Patents page), which is promising for bringing the currently limited and expensive advanced ultrasound elastography technique to the everyday life of patients.
Prior to MIT, I studied at Tsinghua University in Beijing, China. I was interested in using mathematical model to explain interesting phenomena in nature: how different animals drink and how they fly, and apply learnings from nature to engineering designs. Check out the Publications page for more detailed information.
Besides research, I enjoy running, working out, movies, music, photography, hiking and travelling to keep a work-life balance. I plan to share some of the images I photographed on my webpage soon!
[Last edit: 04/20/2022]