EvolvingGrasp: Evolutionary Grasp Generation via Efficient Preference Alignment

2025-03-18

AI Summary

To address the limited generalization capability of dexterous hands in open, complex environments—stemming from insufficient diversity in training data—this paper proposes an evolutionary grasp generation framework based on preference alignment. Our method integrates preference learning, reinforcement learning, and high-fidelity physics simulation. Key contributions include: (1) Handpose-wise Preference Optimization (HPO), the first fine-grained approach to modeling grasp pose preferences; and (2) a physics-aware consistency model that jointly enforces simulated dynamics constraints and lightweight consistency distillation, ensuring both physical feasibility and real-time inference. Evaluated on four major benchmarks—YCB-Video, DexYCB, EPIC-Kitchens, and Real2Sim—our method achieves state-of-the-art grasp success rates and sampling efficiency. Moreover, it demonstrates strong sim-to-real generalization across domains, validating robust cross-environment adaptability.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Intelligent Robots: Manipulation
• Machine Learning: Imitation Learning & Inverse Reinforcement Learning

Application Category

• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Web Mining and Content Analysis: Web data generation and simulation
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Dexterous robotic hands often struggle to generalize effectively in complex environments due to the limitations of models trained on low-diversity data. However, the real world presents an inherently unbounded range of scenarios, making it impractical to account for every possible variation. A natural solution is to enable robots learning from experience in complex environments, an approach akin to evolution, where systems improve through continuous feedback, learning from both failures and successes, and iterating toward optimal performance. Motivated by this, we propose EvolvingGrasp, an evolutionary grasp generation method that continuously enhances grasping performance through efficient preference alignment. Specifically, we introduce Handpose wise Preference Optimization (HPO), which allows the model to continuously align with preferences from both positive and negative feedback while progressively refining its grasping strategies. To further enhance efficiency and reliability during online adjustments, we incorporate a Physics-aware Consistency Model within HPO, which accelerates inference, reduces the number of timesteps needed for preference finetuning, and ensures physical plausibility throughout the process. Extensive experiments across four benchmark datasets demonstrate state of the art performance of our method in grasp success rate and sampling efficiency. Our results validate that EvolvingGrasp enables evolutionary grasp generation, ensuring robust, physically feasible, and preference-aligned grasping in both simulation and real scenarios.

Problem

• Improve robotic grasp generalization in complex environments
• Enable continuous learning from feedback for optimal performance
• Enhance grasp efficiency and reliability through physics-aware models

Innovation

• Evolutionary grasp generation via preference alignment
• Handpose wise Preference Optimization for feedback learning
• Physics-aware Consistency Model enhances efficiency, reliability