Ph.D. in Robotics, University of Michigan

Advisor: Dmitry Berenson, Co-advisor: Nima Fazeli

Dissertation: Contact-based Perception and Planning for Robotic Manipulation in Novel Environments

B.ASc. in Engineering Science Robotics, University of Toronto

GPA: 3.914.0, Major GPA: 4.04.0, Class standing 2161 with high honors

S. Zhong, N. Fazeli, and D. Berenson, “RUMI: Rummaging with Mutual Information,” IEEE Transactions on Robotics (T-RO), 2024. (under review)

S. Zhong, N. Fazeli, and D. Berenson, “CHSEL: Producing Diverse Plausible Pose Estimates from Contact and Free Space Data,” Robotics: Science and Systems (RSS), 2023. link

S. Zhong, N. Fazeli, and D. Berenson, “Soft Tracking Using Contacts for Cluttered Objects to Perform Blind Object Retrieval,” IEEE Robotics and Automation Letters (RA-L), vol. 7, no. 2, pp. 3507-3514, April 2022. link

S. Zhong, Z. Zhang, N. Fazeli, and D. Berenson, “TAMPC: An Online Controller for Escaping Traps in Novel Environments,” IEEE Robotics and Automation Letters (RA-L), vol. 6, no. 2, pp. 1447-1454, April 2021. link

K. E. Murray, O. Petelin, S. Zhong, J. M. Wang et al., “Vtr 8: High-performance CAD and Customizable FPGA Architecture Modelling,” ACM Transactions on Reconfigurable Technology and Systems (TRETS), vol. 13, no. 2, pp. 1-55, 2020. Best Paper Award. link

PyTorch Volumetric Data Structures

github.com/UM-ARM-Lab/pytorch_volumetric pythonPyTorch

Signed distance functions (SDF), voxel grids, and other volumetric data structures accelerated on the GPU.

Fast parallel query across points for SDF value and gradients

PyTorch Differentiable Robot Kinematics

github.com/UM-ARM-Lab/pytorch_kinematics pythonPyTorch

Parallel and differentiable robot forward kinematics and Jacobian calculation.

Load robot description from URDF, SDF, and MJCF formats

PyTorch Model Predictive Path Integral Controller

github.com/UM-ARM-Lab/pytorch_mppi pythonPyTorch

Batched and GPU accelerated Model Predictive Path Integral (MPPI) controller.

Novel kernel interpolation smoothing

Senior Applied Scientist, Simulation at Cruise

Cruise works on vehicle autonomy.

Extended internal VLM framework to handle simulation data

Verity Studios R&D Engineering Intern with Prof. Raffaello D'Andrea

16 months Professional Experience Year, Zurich veritystudios.com C++QTboost

Modelled novel drone indoors localization system using physics first principles

Achieved correlation of 0.86 (95% confidence >0.80) against experimental performance

Designed and developed parameter server firmware used across different hardware platforms

FPGA CAD Routing Optimization with Prof. Vaughn Betz

Summer research with USRA NSERC 5k grant, University of Toronto johnsonzhong.me/projects/vpr C++

Verilog-to-Routing (VTR) is a CAD flow mapping Verilog to FPGAs. Its runtime performance was bottlenecked by the routing phase for large circuits.

Developed route tree pruning algorithm to allow incremental reroutes, speeding up routing by up to 3x on difficult benchmarks

Designed targeted rerouting algorithm for critical yet suboptimal connections, producing up to 30% faster resulting circuits (maximum frequency)

Benchmarked over realistic circuits, with speedups scaling with circuit size

Engineering Science Award of Excellence johnsonzhong.me/res/grad/award_of_excellence.pdf

3rd in Ontario Engineering Competition 2018 Programming category 500

1st in Ontario Engineering 2016 Competition Programming category 2000 johnsonzhong.me/projects/snowfun

1st in Canada in IEEEXtreme 9.0 (286800 globally) johnsonzhong.me/res/ieee9.pdf

Context.io API prize in PennApps Winter 2015 500 devpost.com/software/snowball

8th in Canada in IEEEXtreme 8.0 (526500 globally) johnsonzhong.me/res/ieee8.pdf

Google Cloud Platform prize in Hack the North 2015 1000 devpost.com/software/forenships

6th in Canada in IEEEXtreme 7.0 (437500 globally) johnsonzhong.me/res/ieee.jpg