Visual Goal-Directed Meta-Imitation Learning

Rivera, Corban G.; Handelman, David A.; Ratto, Christopher R.; Patrone, D.; Paulhamus, Bart L.

doi:10.1109/cvprw56347.2022.00421

Cited by 2 publications

(1 citation statement)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Compared to the RL agent, nonetheless, the human being uses a meta-heuristic approach when the given task is highly complex, such as dexterous hand manipulation [32,36]. Existing studies have reported successful results for placing and pushing [33,37], door-locking, and Jenga [38]; however, the feasibility of the RL algorithm for learning human-like meta-heuristic behavior in dexterous hand manipulation has not yet been fully accounted for.…”

Section: Introductionmentioning

confidence: 99%

Solving a Simple Geduldspiele Cube with a Robotic Gripper via Sim-to-Real Transfer

et al. 2022

View full text Add to dashboard Cite

Geduldspiele cubes (also known as patience cubes in English) are interesting problems to solve with robotic systems on the basis of machine learning approaches. Generally, highly dexterous hand and finger movement is required to solve them. In this paper, we propose a reinforcement-learning-based approach to solve simple geduldspiele cubes of a flat plane, a convex plane, and a concave plane. The key idea of the proposed approach is that we adopt a sim-to-real framework in which a robotic agent is virtually trained in simulation environment based on reinforcement learning, then the virtually trained robotic agent is deployed into a physical robotic system and evaluated for tasks in the real world. We developed a test bed which consists of a dual-arm robot with a patience cube in a gripper and the virtual avatar system to be trained in the simulation world. The experimental results showed that the virtually trained robotic agent was able to solve simple patience cubes in the real world as well. Based on the results, we could expect to solve the more complex patience cubes by augmenting the proposed approach with versatile reinforcement learning algorithms.

show abstract

Section: Introductionmentioning

confidence: 99%

Solving a Simple Geduldspiele Cube with a Robotic Gripper via Sim-to-Real Transfer

et al. 2022

View full text Add to dashboard Cite

show abstract

Autonomous Manipulation Learning for Similar Deformable Objects via Only One Demonstration

Ren,

Chen,

Cong

2023

2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

View full text Add to dashboard Cite

Learning robust and generalizable manipulation skills from demonstrations remains a key challenge in robotics, with broad applications in industrial automation and service robotics. While recent imitation learning methods have achieved impressive results, they often require large amounts of demonstration data and struggle to generalize across different spatial variants. In this work, we present a novel framework that learns manipulation skills from as few as 10 demonstrations, yet still generalizes to spatial variants such as different initial object positions and camera viewpoints. Our framework consists of two key modules: Semantic Guided Perception (SGP), which constructs task-focused, spatially aware 3D point cloud representations from RGB-D inputs; and Spatial Generalized Decision (SGD), an efficient diffusion-based decision-making module that generates actions via denoising. To effectively learn generalization ability from limited data, we introduce a critical spatially equivariant training strategy that captures the spatial knowledge embedded in expert demonstrations. We validate our framework through extensive experiments on both simulation benchmarks and real-world robotic systems. Our method demonstrates a 60-70% improvement in success rates over state-of-the-art approaches on a series of challenging tasks, even with substantial variations in object poses and camera viewpoints. This work shows significant potential for advancing efficient, generalizable manipulation skill learning in real-world applications. 1

show abstract

Visual Goal-Directed Meta-Imitation Learning

Cited by 2 publications

References 13 publications

Solving a Simple Geduldspiele Cube with a Robotic Gripper via Sim-to-Real Transfer

Solving a Simple Geduldspiele Cube with a Robotic Gripper via Sim-to-Real Transfer

Autonomous Manipulation Learning for Similar Deformable Objects via Only One Demonstration

Contact Info

Product

Resources

About