Adversarial Game-Theoretic Algorithm for Dexterous Grasp Synthesis

2025-11-08

AI Summary

Existing grasp synthesis methods for multi-fingered robotic hands optimize only against single external disturbances, neglecting adversarial object motions—such as active escape maneuvers—leading to poor robustness under dynamic perturbations. Method: This work formulates grasp generation as a zero-sum game between the robot and the object: the robot optimizes for stable grasp configurations while the object concurrently seeks its optimal escape trajectory. Our approach integrates kinematic constraint-aware optimization with an efficient adversarial search strategy, explicitly modeling dynamic interaction while maintaining real-time feasibility. Results: In simulation, our method achieves a mean success rate of 75.78%, outperforming the state-of-the-art by 19.61%. On physical platforms—ShadowHand and LeapHand—it attains success rates of 85.0% and 87.5%, respectively. These results demonstrate substantial improvements in robustness against both environmental disturbances and intentional adversarial object motion.

Technology Category

• Intelligent Robots: Manipulation
• Multiagent Systems: Adversarial Agents
• Machine Learning: Adversarial Learning & Robustness

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: Uses of LLMs and GenAI for marketplace design, bidding, and strategic interactions

Abstract

For many complex tasks, multi-finger robot hands are poised to revolutionize how we interact with the world, but reliably grasping objects remains a significant challenge. We focus on the problem of synthesizing grasps for multi-finger robot hands that, given a target object's geometry and pose, computes a hand configuration. Existing approaches often struggle to produce reliable grasps that sufficiently constrain object motion, leading to instability under disturbances and failed grasps. A key reason is that during grasp generation, they typically focus on resisting a single wrench, while ignoring the object's potential for adversarial movements, such as escaping. We propose a new grasp-synthesis approach that explicitly captures and leverages the adversarial object motion in grasp generation by formulating the problem as a two-player game. One player controls the robot to generate feasible grasp configurations, while the other adversarially controls the object to seek motions that attempt to escape from the grasp. Simulation experiments on various robot platforms and target objects show that our approach achieves a success rate of 75.78%, up to 19.61% higher than the state-of-the-art baseline. The two-player game mechanism improves the grasping success rate by 27.40% over the method without the game formulation. Our approach requires only 0.28-1.04 seconds on average to generate a grasp configuration, depending on the robot platform, making it suitable for real-world deployment. In real-world experiments, our approach achieves an average success rate of 85.0% on ShadowHand and 87.5% on LeapHand, which confirms its feasibility and effectiveness in real robot setups.

Problem

• Synthesizing reliable multi-finger robot grasps that resist object motion
• Addressing object escape potential ignored by single-wrench focused approaches
• Formulating grasp synthesis as adversarial game between robot and object

Innovation

• Adversarial game-theoretic grasp synthesis for multi-finger hands
• Two-player game formulation between robot and object
• Real-time grasp generation with improved success rates