Contact implicit admm finds multiple ways to handle contact tasks
Contact-Implicit Stein Projected ADMM for Discovery of Diverse Contact-Rich Manipulation Strategies
Robotics
Summary
Robots often need to touch and manipulate objects in many ways, but existing methods usually find just one way and get stuck there. The authors designed a new technique that helps robots discover many different ways to touch and move objects at once, making their behavior more flexible and reliable. They combined a known optimization approach with a method to keep solutions diverse and tested it on tasks like pushing, grasping, and handing over objects between robots. This approach finds multiple useful contact strategies more easily than older methods.
What this means in practice
- •For robotics engineers: Enable design of manipulation algorithms that can flexibly choose among multiple contact strategies for tasks like grasping and pushing.
- •For manufacturing automation teams: Improve robot adaptability by generating diverse handling methods for contact-intensive assembly and part handling.
Authors
Hrishikesh Sathyanarayan, Christian Hughes, Ian Abraham
Abstract
Contact-implicit trajectory optimization formulates contact-rich manipulation as a single constrained program; however, that single program run collapses onto one local optimum out of many equally valid contact modes, grasps, or push directions. As a consequence, the resulting manipulation strategy is reluctant to change and sensitive to initialization. In order to promote robust manipulation, this paper investigates how contact-implicit solvers can discover diverse contact-rich strategies. Our approach derives a variation of Consensus Alternating Direction Method of Multipliers (ADMM) combined with Stein variational inference methods to output a set of distinct contact-rich solutions. We find that applying the Stein repulsive force to ADMM's split variable (rather than its primal form) allows for effective coverage over the set of feasible contact strategies without prematurely stalling the solver. We demonstrate the effectiveness of our approach on a variety of contact-rich manipulation tasks, including pushing, grasping, and multi-robot handover. Last, we find the proposed solver is simpler in form and capable of discovering unique contact modes when compared with existing solvers. Videos and code with examples are found in https://anon-website-submission.github.io/stein-admm-website/.