Hanna Zhang

I am a Graduate Technology Intern at Disney Research within Walt Disney Imagineering, where I work on next generation performing stunt robots.

Prior to Disney, I finshed my Masters in Computer Science at the University of Toronto under Jessica Burgner-Kahrs. There I worked on convex relaxation-based solving techniques for continuum robot inverse kinematics. I have a Bachelors of Applied Science in Engineering Science majoring in Robotics minoring in Bioengineering from University of Toronto.

Email  /  GitHub  /  Google Scholar  /  LinkedIn  /  CV

Research

I'm interested in control and planning for robots.

CIDGIKc: Distance-Geometric Inverse Kinematics for Continuum Robots

We present CIDGIKc (Convex Iteration for Distance-Geometric Inverse Kinematics for Continuum Robots), an algorithm that solves these nonconvex feasibility problems with a sequence of semidefinite programs whose objectives are designed to encourage low-rank minimizers. CIDGIKc is enabled by a novel distance-geometric parameterization of constant curvature segment geometry for CRs with extensible segments.

Cooperative Control of Dual-Arm Concentric Tube Continuum Robots

Many real-world tasks often require collaboration of multiple manipulators necessitating dual-arm (DA) robotic systems. Leveraging the benefits of CTCRs in a dual-arm configuration, i.e. DA-CTCR, has great potential for applications in such areas. A hierarchy-based control framework is proposed to efficiently control and coordinate motions in a semi-autonomous manner for DACTCRs. This framework is validated in various assistive/motion coordination control schemes in simulation and on a real-robot prototype.

Semi-permanent tattoos, patent for novel application technology, semi-permanent ink, and manufacturing process

This patent relates to semi-permanent tattoos, and more particularly to genipin-based ink compositions as well as layered adhesive articles and methods for application of the genipin-based, semi-permanent tattoos. This encompasses work I did as a year long research team intern at Inkbox Inc.

Technical Projects

Here is a collection of interesting projects I've worked on.

Towards Improved Modelling and Classification of Mobile-Recorded Cough Audio for Diagnosis of Respiratory Conditions: A Generative Model Approach

With 2 fellow graduate students I evaluated the ability of our conditioned generative model to produce wet or dry and female or male coughs by running statistical tests on the differences in MFCCs (Mel-frequency Cepstral Coefficients), a standard audio feature metric, and model embedding on generated coughs.

Linear-Time Dynamics using Lagrange Multipliers

Worked with a fellow student to implement David Baraff’s 1996 work on Linear-Time Dynamics using Lagrange Multipliers in MATLAB. Results are demonstrated on 2D serially jointed structures, trees, and closed-loop chain structures. The dynamics of these systems are handled by 3 Different methods 1) standard matrix inversion, 2) a sparse O(n) factorization method, and a 3) dense \textit{O(n$^3$)} factorization method. Comparisons of the performance of these different solvers on the aforementioned systems at varying scales (i.e. size of the structures) were done.

Generation of a Combined Bird’s Eye View of a Region around a Robotic Rover From Omnidirectional Stereo Images

Generated meaningful bird’s eye view images from 360\textdegree omnidirectional images taken on a small robotic rover travelling along a path through simulated Mars/Moon terrain. Reconstructed a flat orthographic map of a region around the rover at a particular time point using omnidirectional images and 3-D point clouds of the terrain. Corrected for variable auto-exposure of the multiple cameras (due to sun position) to improve the readability of that image.

Leadership

Some opportunities I got to organize and lead meaningful causes and events.

Vice Chair Operations: University of Toronto Engineering Orientation Committee

Led a team of over 400 volunteers to deliver a week of online welcoming activities as well as some in-person programming for over 1000 incoming first-year engineering students, with a budget over $130k. Adapted/re-conceptualized Orientation activities which included new initiatives to encourage community engagement, make it easy for people to make genuine connections with one another, and offer a memorable, positive welcome into their first year of university. Oversaw the assembly/shipping/delivery logistics of 800+ U of T Engineering kits with swag, hosted international in-person meetups complying with safety regulations under COVID-19, ran online outreach/social initiatives, organized online study communities, and supported the re-design of orientation leader training.

Design and source code from Jon Barron's website