Our Work

Here shows our work related to Robot Learning.

background-rl.png

Research Overview

Robot learning techniques, such as reinforcement learning and imitation learning, will be investigated for the development of intelligent robots. To improve the efficiency of training a reinforcement learning model, effective representation learning and multi-modal sensor fusion algorithms will be explored. To enable the robots to learn unknown dynamics and acquire complex visual-motor skills via visual observation with generalizability, one/few-shot imitation learning will be investigated for the robots to learn from a small dataset of human demonstration.

In addition, Explainable AI will be developed to eliminate the black-box effects for deep learning-based algorithms, which can enhance human’ trust when interacting with robots. Applications include domestic robots, industrial robots, medical robots and warehouse robots.  

XAI for Intelligent Robotics

XAI.png

Target

  • Enhance User's Trust with Explainable Interface

  • Enhance User's Trust with Model Explanation

  • Enhance User's Trust with Uncertainty Prediction

  • Enhance User's Trust by Tracing Failures

01

Explainable Hierarchical Imitation Learning

Dandan Zhang; Qiang Li; Yu Zheng; Dongsheng Zhang; Zhengyou Zhang

EHIL.png

To accurately pour drinks into various containers is an essential skill for service robots. However, drink pouring is a dynamic process and difficult to model. 
Traditional deep imitation learning techniques for implementing autonomous robotic pouring have an inherent black-box effect and require a large amount of demonstration data for model training. To address these issues, an Explainable Hierarchical Imitation Learning (EHIL) method is proposed in this paper such that a robot can learn high-level general knowledge and execute low-level actions across multiple drink pouring scenarios. Moreover, with the EHIL method, a logical graph can be constructed for task execution, through which the decision-making process for action generation can be made explainable to users and the causes of failure can be traced out. Based on the logical graph, the framework is manipulable to achieve different targets while the adaptability to unseen scenarios can be achieved in an explainable manner.
A series of experiments have been conducted to verify the effectiveness of the proposed method. Results indicate that EHIL outperforms the traditional behavior cloning method in terms of success rate, adaptability, manipulability and explainability. 

Full paper link:

https://ieeexplore.ieee.org/document/9667114

logicGraph.png
tracefailures.png

02

One-Shot Domain-Adaptive Imitation Learning
via Progressive Learning

Dandan Zhang; Wen Fan; John Llyod; Chenguang Yang;Nathan Lepora

oneshot-progressive.png

Traditional deep learning-based visual imitation learning techniques require a large amount of demonstration data for model training, and the pre-trained models are difficult to adapt to new scenarios. To address these limitations,
we propose a unified framework using a novel progressive learning approach comprised of three phases: i) a coarse learning phase for concept representation, ii) a fine learning phase for action generation, and iii) an imaginary learning phase for domain adaptation. Overall, this approach leads to a one-shot domain-adaptive imitation learning framework. We use robotic pouring task as an example to evaluate its effectiveness. Our results show that the method has several advantages over contemporary end-to-end imitation learning approaches, including an improved success rate for task execution and more efficient training for deep imitation learning. In addition, the generalizability to new domains is improved, as demonstrated here with novel background, target container and granule combinations. We believe that the proposed method can be broadly applicable to different industrial or domestic applications that involve deep imitation learning for robotic manipulation, where the target scenarios have high diversity while the human demonstration data is limited.

Full paper link:

https://arxiv.org/abs/2204.11251

•A. Explainable Intelligent Robotics

•B. Domain Adaptation for Robotic Manipulation

•C. Data-Efficient Robot Learning for Robotic Manipulation

•D. Continual Learning for Robotic Manipulation

•E. Sim-to-Real Transfer Learning for Robotic Manipulation

•F. Multi-modality Representation Learning for Contact-Rich Task

Ongoing Projects