Brandon H. Meng
Ph.D., Computer & Information Sciences · DePaul University
I develop sampling- and graph-based motion planning algorithms for high-dimensional robots in dynamic environments, where plans must respect the robot's kinematics, the timing of moving obstacles, and uncertainty. My dissertation addresses path planning for continuum arms in static and dynamic settings. I am extending this work toward multi-manipulator planning and toward integrating learning with model-based methods. I am seeking faculty and postdoctoral positions and, as a U.S. and German citizen, I can work in the US or EU without visa sponsorship.
Education
-
2019 – 2026
Ph.D., Computer & Information Sciences
DePaul University · Advisor: Iyad Kanj · GAANN Fellow
Dissertation: Path Planning for Continuum Arms · PDF
- Path planners for multisection continuum arms
- Planners that anticipate moving obstacles, extended to dual‑arm systems
- Codebase written from scratch in C++ and Python, including simulators and a Three.js visualizer
-
2018 – 2019
M.S., Computer Science
DePaul University · Advisor: Iyad Kanj
Thesis: Machine-Learning Based Approach for Efficient Path Planning of Continuum Co‑Robotic Arms
-
2014 – 2018
B.S., Computer Science
DePaul University
Software Development concentration; Mathematical Sciences minor
Experience
-
2024 – 2026
Adjunct Faculty
School of Computing, DePaul University
Instructor of record for 6 sections of CSC 241/242, Introduction to Computer Science I & II
-
2021, 2023
Assistant Lecturer
School of Computing, DePaul University
CSC 321, Design and Analysis of Algorithms (2 sections)
-
2018 – 2019
Graduate & Undergraduate Tutor
Computer Science, DePaul University
-
2017, 2018
Summer Intern
United Airlines, Chicago
Ansible automation, AWS
-
2016
Research Intern
Discovery Lab Global, Columbus, OH
Service
Peer reviewer for
- IEEE RA-L
- ICRA
- IEEE T-ASE
- IROS
- RoboSoft
Awards
- GAANN Fellowship 2019 – 2024
- Presidential Scholarship 2014 – 2018
Skills
- Languages C++, Python, JavaScript, Java, SQL
- Libraries OMPL, CGAL, Boost, Eigen, NLopt, NumPy, SciPy, PyTorch, OpenCV
- Tools Three.js, MATLAB, Blender
Publications
-
RA-L 2022
RRT*-Based Path Planning for Continuum Arms
doi:10.1109/LRA.2022.3174257 -
RoboSoft 2024
Path Planning for Continuum Arms in Dynamic Environments
doi:10.1109/RoboSoft60065.2024.10521950 -
ICRA 2021
Anticipatory Path Planning for Continuum Arms in Dynamic Environments
doi:10.1109/ICRA48506.2021.9560952 -
ICRA 2021
Smooth Path Planning for Continuum Arms
doi:10.1109/ICRA48506.2021.9560982 -
RoboSoft 2019
Near-optimal Smooth Path Planning for Multisection Continuum Arms
doi:10.1109/ROBOSOFT.2019.8722778 -
Under review
Anticipatory Joint-space Planning for Continuum Arms in Dynamic Environments
-
In preparation
Dual-Arm Path Planning for Continuum Manipulators
Talks
- 2024 Path Planning for Continuum Arms in Dynamic Environments — IEEE RoboSoft, San Diego
- 2022 RRT*-Based Path Planning for Continuum Arms — IEEE IROS, Kyoto
- 2021 Anticipatory Path Planning for Continuum Arms — IEEE ICRA, Xi’an
- 2021 Smooth Path Planning for Continuum Arms — IEEE ICRA, Xi’an
Live demos
Check out live demos of planners from my work. A three-section continuum arm runs entirely in your browser: set a target for the tip, add random obstacles, choose a planner and press Plan to watch it search.
-
W-Space RRT*
An RRT* grown over tip positions, extended with weighted Jacobian steps and null-space motion.
From RRT*-Based Path Planning for Continuum Arms (RA-L 2022)
-
Look-up table
Precomputed configurations binned into workspace cubes, searched as a cube graph.
From Anticipatory Path Planning for Continuum Arms in Dynamic Environments (ICRA 2021)
-
Joint-space PRM
A roadmap in tendon space with inverse-kinematics goals and lazy collision checking.
From Anticipatory Joint-space Planning for Continuum Arms in Dynamic Environments (RA-L, under review)
The demos are the static versions of these planners; the moving-obstacle parts are not included.
Contact
Chicago, IL ·