ILBiT: Imitation Learning for Robot Using Position and Torque Information based on Bilateral Control with Transformer

Osaka University

Masato Kobayashi*, Thanpimon Buamanee, Yuki Uranishi, Haruo Takemura

ILBiT

ILBiT
ILBiT: Imitation Learning for Robot Using Position and Torque Information based on Bilateral Control with Transformer, arXiv, 2024.
Abstract:

Autonomous manipulation in robot arms is a complex and evolving field of study in robotics. This paper introduces an innovative approach to this challenge by focusing on imitation learning (IL). Unlike traditional imitation methods, our approach uses IL based on bilateral control, allowing for more precise and adaptable robot movements. The conventional IL based on bilateral control method have relied on Long Short-Term Memory (LSTM) networks. In this paper, we present the IL for robot using position and torque information based on Bilateral control with Transformer (ILBiT). This proposed method employs the Transformer Encoder model, known for its robust performance in handling diverse datasets and its capability to surpass LSTM’s limitations, especially in tasks requiring detailed force adjustments. A standout feature of ILBiT is its high-frequency operation at 100 Hz, which significantly improves the system’s adaptability and response to varying environments and objects of different hardness levels. The effectiveness of the Transformer-based ILBiT method can be seen through comprehensive real-world experiments.

Overview

ILBiT
Step1: Data Collection via Bilateral Control, Step2: Transformer Encoder Model for Learning and Prediction, Step3: Execution through Trained Model
Data Collection via Bilateral Control (Step 1)

The first step involves the collection of demonstrative data for IL via bilateral control.

Transformer Model for Learning and Prediction (Step 2)

Step 2 employs the Transformer Encoder model to process the collected data. This model excels in handling diverse datasets and is adept at learning complex patterns of human manipulation, enabling the system to predict and modulate forces accurately for effective robotic manipulation.

Execution through Trained Model (Step 3)

The system autonomously performing tasks with a robotic arm using the Transformer Encoder, showcasing adaptability and efficiency in real-world settings.

Citation


@misc{kobayashi2024ilbit,
    title={ILBiT: Imitation Learning for Robot Using Position and Torque Information based on Bilateral Control with Transformer},
    author={Masato Kobayashi and Thanpimon Buamanee and Yuki Uranishi and Haruo Takemura},
    year={2024},
    eprint={2401.16653},
    archivePrefix={arXiv},
    primaryClass={cs.RO}
}

Contact

Masato Kobayashi (Assistant Professor, Osaka University, Japan)