Approximating Global Contact-Implicit MPC via Sampling and Local Complementarity

2025-05-19

AI Summary

Global optimization of contact sequences in non-prehensile dexterous manipulation remains challenging; existing methods are constrained by local interactions and lack real-time solvability. Method: This paper proposes a real-time Contact-Implicit Model Predictive Control (CI-MPC) framework. Its core innovation introduces a pre-contact free-motion phase within each control cycle, integrating low-dimensional global end-effector pose sampling with parallel local contact-aware MPC cost evaluation—enabling, for the first time, tightly coupled global exploration and local precision optimization. The framework unifies contact-implicit dynamics modeling, complementarity-constrained optimization, and receding-horizon planning. Results: Validated on a Franka Panda platform, the method generates diverse, real-time contact sequences, successfully manipulating non-convex objects. It significantly improves task success rate and policy robustness compared to prior approaches.

Technology Category

• Intelligent Robots: Manipulation
• Search and Optimization: Mixed Discrete/Continuous Search
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems

Abstract

To achieve general-purpose dexterous manipulation, robots must rapidly devise and execute contact-rich behaviors. Existing model-based controllers are incapable of globally optimizing in real-time over the exponential number of possible contact sequences. Instead, recent progress in contact-implicit control has leveraged simpler models that, while still hybrid, make local approximations. However, the use of local models inherently limits the controller to only exploit nearby interactions, potentially requiring intervention to richly explore the space of possible contacts. We present a novel approach which leverages the strengths of local complementarity-based control in combination with low-dimensional, but global, sampling of possible end-effector locations. Our key insight is to consider a contact-free stage preceding a contact-rich stage at every control loop. Our algorithm, in parallel, samples end effector locations to which the contact-free stage can move the robot, then considers the cost predicted by contact-rich MPC local to each sampled location. The result is a globally-informed, contact-implicit controller capable of real-time dexterous manipulation. We demonstrate our controller on precise, non-prehensile manipulation of non-convex objects using a Franka Panda arm. Project page: https://approximating-global-ci-mpc.github.io

Problem

• Optimizing real-time contact-rich robot manipulation globally
• Overcoming limitations of local contact-implicit control models
• Enabling dexterous manipulation via global sampling and local MPC

Innovation

• Combines local complementarity control with global sampling
• Uses contact-free stage before contact-rich stage
• Real-time dexterous manipulation via global-informed MPC