| CARVIEW |
Select Language
HTTP/2 200
server: GitHub.com
content-type: text/html; charset=utf-8
last-modified: Tue, 25 Nov 2025 22:27:07 GMT
access-control-allow-origin: *
etag: W/"69262d3b-a036"
expires: Mon, 29 Dec 2025 12:30:37 GMT
cache-control: max-age=600
content-encoding: gzip
x-proxy-cache: MISS
x-github-request-id: CA7A:15317B:8CCC1E:9E09E9:69527215
accept-ranges: bytes
age: 0
date: Mon, 29 Dec 2025 12:20:37 GMT
via: 1.1 varnish
x-served-by: cache-bom-vanm7210030-BOM
x-cache: MISS
x-cache-hits: 0
x-timer: S1767010837.488365,VS0,VE213
vary: Accept-Encoding
x-fastly-request-id: 85edfb1af8169ea18d2d72daf77967d2171236b7
content-length: 7428
Mark Van der Merwe
IEEE ICRA 2025
Biography
I am a PhD student in Robotics at the University of Michigan, advised by Dmitry Berenson and Nima Fazeli. My research is at the intersections of perception and planning for contact-rich manipulation. I also spent time at MERL working with Devesh Jha.
Before joining Michigan, I was an Intern at NASA Jet Propulsion Laboratory working on the Perception Systems team. I worked on the proposed Mars Sample Return project. I graduated with my Honors Bachelor’s Degree in Computer Science from the University of Utah in May 2020, where I worked with Tucker Hermans on geometrically-aware robotic grasping (see our ICRA 2020 paper). During my undergraduate degree, I also interned at Mila under the supervision of Liam Paull.
News
- Paper accepted at CoRL 2025 (here).
- Paper accepted at RA-L 2025 (here). I will be presenting this work at the CRM workshop at ICRA 2025!
- Paper accepted at ICRA 2025 (here)
Featured Publications
In-Context Iterative Policy Improvement for Dynamic Manipulation
Attention-based architectures trained on internet-scale language data have demonstrated state of the art reasoning ability for various …
CoRL 2025
Estimating Deformable-Rigid Contact Interactions for a Deformable Tool via Learning and Model-Based Optimization
Dexterous manipulation requires careful reasoning over extrinsic contacts. The prevalence of deforming tools in human environments, the …
IEEE RA-L 2025
This&That: Language-Gesture Controlled Video Generation for Robot Planning
We propose a robot learning method for communicating, planning, and executing a wide range of tasks, dubbed This&That. We achieve …
Publications
Estimating Deformable-Rigid Contact Interactions for a Deformable Tool via Learning and Model-Based Optimization
IEEE RA-L 2025
This&That: Language-Gesture Controlled Video Generation for Robot Planning
IEEE ICRA 2025
Learning the Dynamics of Compliant Tool-Environment Interaction for Visuo-Tactile Contact Servoing
CoRL 2022
Rover Relocalization for Mars Sample Return by Virtual Template Synthesis and Matching
IEEE RA-L 2021
Machine Vision Based Sample-Tube Localization for Mars Sample Return
IEEE Aero 2021
Multi-Fingered Active Grasp Learning
IEEE IROS 2020