CoRL 2026 Workshop

Open Problems in Contact-Rich Loco-Manipulation

How should robots choose, sense, and control contact while moving through the world?

November 12, 20262:00–6:00 PM CST

JW Marriott Austin

Workshop Theme

Recent advances in robot learning, control, perception, and planning have enabled mobile robots to navigate complex environments and interact more capably with the physical world. Legged robots, humanoids, and mobile manipulators now perform tasks that require whole-body motion, forceful interaction, and adaptation under uncertainty.

Many real-world loco-manipulation problems cannot be solved by grasping alone. Robots must combine prehensile skills, such as grasping and carrying, with non-prehensile skills, such as pushing, dragging, pivoting, bracing, or using sustained body-object contact. The contact strategy is directly coupled to locomotion, balance, onboard sensing, planning, and whole-body control.

This workshop brings together researchers in legged locomotion, mobile manipulation, non-prehensile manipulation, whole-body control, contact-aware planning, onboard perception, and sim-to-real transfer to identify shared challenges and promising research directions.

Core Challenges

Contact-mode decision making

Robots must decide when to grasp, push, drag, brace, lean, pin, wedge, pivot, or exploit the environment, and how to transition between those strategies under uncertainty.

Whole-body coordination under contact

Contacts involving a robot's arms, legs, torso, or base, as well as objects and the environment, affect balance, mobility, force regulation, and task success.

Reasoning under uncertainty

Robots must act despite uncertainty in friction, mass, compliance, center of mass, contact location, slip, object geometry, terrain, and other hidden state.

Generalization across embodiments

The field needs principles and representations that transfer across fixed-base arms, mobile manipulators, quadrupeds with arms, humanoids, soft robots, and dexterous hands.

Benchmarks, metrics, and infrastructure

Progress requires shared task families, metrics, simulation assets, and real-world protocols that capture robustness, safety, recoverability, contact efficiency, and generalization.

Invited Speakers

Emo Todorov

University of Washington Roboti

Resurrecting Model-Based Control and Learning

Watch the talk · 35:39
0:00 / 0:00

Tao Pang

Multiply Labs

How can we efficiently leverage coding agents for dexterous manipulation?

Watch the talk · 23:13
0:00 / 0:00

Student Talks

Presented by

Workshop Format

Full schedule

Invited talks

Four invited talks will frame the central open problems and leave time for direct Q&A on contact-mode selection, whole-body coordination, and uncertainty.

Lightning talks and posters

Four selected papers will be introduced through five-minute spotlight talks. Accepted workshop papers will be discussed in an interactive poster session.

Panel and community report

A moderated panel will bring together speakers, authors, and attendees, with key open problems and research directions summarized in a post-workshop report.

Workshop Moments

Thank you for joining us

See you next time.

JW Marriott Austin · November 12, 2026