Seth Isaacson

sethgi (AT) umich (DOT) edu

I'm a Ph.D. candidate in Robotics at the University of Michigan. I'm working to improve robots' abilities to reliably understand and navigate complex, cluttered, and human-centric environments.

Research

PhD Research

Let's Make a Splan: Risk-Aware Trajectory Optimization in a Normalized Gaussian Splat

Jonathan Michaux*, Seth Isaacson*, Challen Enninful Adu, Adam Li, Rahul Kashyap Swayampakula, Parker Ewen, Sean Rice, Katherine A. Skinner and Ram Vasudevan.

[Paper] [Project Website]
Splanning Figure 1

SAD-GS: Shape-aligned Depth-supervised Gaussian Splatting

Pou-Chun Kung, Seth Isaacson, Ram Vasudevan, and Katherine A. Skinner.

[Paper] [Code] [Project Website]
Splanning Figure 1

[RA-L 2023 Best Paper] LONER: LiDAR Only Neural Representations for Real-Time SLAM

Seth Isaacson*, Pou-Chun Kung*, Mani Ramanagopal, Ram Vasudevan, and Katherine A. Skinner.

[Paper] [Code] [Project Website]
LONER Figure 1

Undergrad Research

MuddSub

[Project Website]
MuddSub

MAD-TN: A Tool for Evaluating Fluency in Human-Robot Interaction

Seth Isaacson*, Gretchen Rice*, and James C. Boerkoel Jr.

[Paper]
MAD-TN

Experience

Optimus Ride / Magna Advanced Development Center

Software Engineer - Mapping and Sensor Systems
  • Designed and implemented of C++ visual mapping pipeline for low-speed ADAS.
  • Wrote sensor interface drivers and conduct evaluations of new sensors (LIDAR, Radar, INS, and GPS) to improve the capabilities of Level 4 autonomous shuttles
  • Contributed high-quality C++ code to the autonomy stack. Participated in code reviews and engaged in discussions on how to improve software development processes.
August 2021 - August 2022

Argo AI

Software Engineering Intern - Onboard Accelerators
  • Researched and prototyped GPU scheduling approaches to make the timing of GPU computations more reliable and increase determinism of the pipeline.
  • Demonstrated that the selected approach effectively eliminates variance in timing, making safety-critical aspects of the system more reliable.
  • Selected to present work in international all-hands meeting with the president and CEO to 1100+ person company.
Summer 2020

Rethink Robotics

Engineering Intern
  • Developed and tested new applications and end-of-arm-tooling for manufacturing robots to allow the robots to be sold into new markets.
  • Investigated implementation of machine learning classifiers to detect when the robot had succeeded or failed at a task.
Summer 2017

Education

Ph.D. in Robotics

University of Michigan, Ann Arbor
Expected May 2027

B.S. in Joint Mathematics and Computer Science

Harvey Mudd College, Claremont CA
May 2021