TY - JOUR
T1 - A Code-Free Interactive Task Programming Interface for Robot Skill Construction
AU - Zhang, Ning
AU - Zhao, Yongjia
AU - Dai, Shuling
N1 - Publisher Copyright:
© King Fahd University of Petroleum & Minerals 2024.
PY - 2025/8
Y1 - 2025/8
N2 - To enhance production efficiency and quality, there is a rising interest in integrating robots into small manufacturing entities (SMEs) to enable flexible and agile production processes, thereby reducing redundancy. This poses challenges for robots as they must perform various tasks in unstructured environments without necessitating specialized programming training for workshop workers. This approach aims to reduce redundancy and streamline operations. To address these challenges, this paper presents a code-free system for programming robot tasks, which leverages an interactive programming interface and kinesthetic teaching methods. The system operates in three phases. Initially, the kinesthetic teaching method is employed to demonstrate the task, and the data generated from this demonstration are utilized for skill acquisition and visualization of the robot’s state within the human–robot interaction software interface. Next, the demonstrated trajectories are learned through dynamic motion primitives, and various sub-skills form the skill library necessary for task completion. Ultimately, the skill library is activated to guide the actual robotic arm in executing the task. This approach allows end users to construct a skill library for a specific task without delving into task code-level programming. To validate the system’s effectiveness, we invited experimenters to utilize our system for programming a designated task. The results demonstrate that users without programming experience can efficiently and flexibly employ our system to program robots for various tasks.
AB - To enhance production efficiency and quality, there is a rising interest in integrating robots into small manufacturing entities (SMEs) to enable flexible and agile production processes, thereby reducing redundancy. This poses challenges for robots as they must perform various tasks in unstructured environments without necessitating specialized programming training for workshop workers. This approach aims to reduce redundancy and streamline operations. To address these challenges, this paper presents a code-free system for programming robot tasks, which leverages an interactive programming interface and kinesthetic teaching methods. The system operates in three phases. Initially, the kinesthetic teaching method is employed to demonstrate the task, and the data generated from this demonstration are utilized for skill acquisition and visualization of the robot’s state within the human–robot interaction software interface. Next, the demonstrated trajectories are learned through dynamic motion primitives, and various sub-skills form the skill library necessary for task completion. Ultimately, the skill library is activated to guide the actual robotic arm in executing the task. This approach allows end users to construct a skill library for a specific task without delving into task code-level programming. To validate the system’s effectiveness, we invited experimenters to utilize our system for programming a designated task. The results demonstrate that users without programming experience can efficiently and flexibly employ our system to program robots for various tasks.
KW - Human–robot interaction
KW - Kinesthetic teaching
KW - Learn from demonstration
KW - Robot skills
UR - https://www.scopus.com/pages/publications/85204426317
U2 - 10.1007/s13369-024-09592-4
DO - 10.1007/s13369-024-09592-4
M3 - 文章
AN - SCOPUS:85204426317
SN - 2193-567X
VL - 50
SP - 11681
EP - 11692
JO - Arabian Journal for Science and Engineering
JF - Arabian Journal for Science and Engineering
IS - 15
ER -