|
Fingertip Force Control based on Contraction Force Model of Elastic Round-belt |
|
Takahiro Inoue(Tokyo Denki University, Japan) |
| This paper presents a novel three-DOF robotic finger that enables fast tapping motion applicable in piano playing and keyboard typing. This study derives force model available for an extended three-DOF finger mechanism that is containing Variable-pitch Screw Module(VpSM) newly designed in this paper. This module is able to eliminate irregular twist phenomenon resulting in unexpected discontinuous movement in finger joints. Thus, we first describe the detail mechanism of the three-DOF robotic finger and indicate great advantage of the finger. Finally, we demonstrate a new force feedback method based on fingertip force estimation without any force/pressure sensors. |
|
|
Exploring Generalization and Adaptability of Offline Reinforcement Learning for Robot Manipulation |
|
Hyun Kim, Injun Park, Ingook Jang, Seonghyun Kim, Samyeul Noh, Joonmyon Cho(ETRI, Korea) |
| This paper empirically investigates offline RL’s feasibility, generalization, and adaptability for robot manipulation tasks compared to online RL approaches. The results demonstrate that offline RL not only shows promising performance but also exhibits better generalization capabilities compared to online RL. In terms of adaptation, offline RL achieves significant performance improvements through small steps of online fine-tuning. These findings underline the potential of offline RL as an effective and practical approach for real-world robot manipulation tasks. |
|
|
Decentralized Fuzzy Control Based Trajectory Tracking of a Robotic Manipulator |
|
Hsiu-Ming Wu, Yi-Yang Cai, Wei-Lun Su(National Taipei University of Technology, Taiwan) |
| The study aims to control a robotic manipulator to achieve the trajectory-tracking of its end-effector. To the objective, a decentralized fuzzy control scheme is proposed. The desired motion trajectories including circular and cardioid shapes of the end-effector are planned such that the corresponding joint limits are mapped to the feasible working space by using the projective inverse kinematics. In the meantime, a common PID control is utilized to compare control performance with the proposed decentralized fuzzy control. Experimental results directly validate that the proposed decentralized fuzzy controller is capable of tracking desired angles of each joint and further achieves trajectory |
|
|
Direct Teaching/Playback Method for the Contact Task of a six-DOF Manipulator using a Parallel-Wire-Type Teaching Device with a Remote Center Compliance Device |
|
Tatsuya Okochi(Nagoya Institute of Technology, Japan), Masao Sakai(Industrial Research Center, Japan), Yoshifumi Morita(Nagoya Institute of Technology, Japan) |
| When performing contact tasks, such as deburring or polishing using a manipulator, a passive compliance is required for the end effector to suppress excessive contact force. Therefore, we developed an end effector that combines a “parallel wire-type teaching device,” which enables direct teaching, and a compliance device. If the stiffness of the end effector is known, it is possible to perform the contact task with the desired pressing force by editing the trajectory of the robot to press against the workpiece. Experiments of feedforward force control on a straight trajectory verified the effectiveness of the proposed teaching method. |
|
|
Experimental Studies of Hybrid Force Control with Time-delayed Compensation for a Robot Manipulator |
|
Jun Woo Lee, Seul Jung(Chungnam National University, Korea) |
| This paper presents the experimental studies of a contact force control task by the hybrid force control with time-delayed control compensation for a robot manipulator. Hybrid force control regulates a desired force directly by minimizing force errors with separating the position and force controlled direction. Integral control is added to the force controller to minimize force errors further. In addition, the time-delayed control scheme is introduced to compensate for uncertainties to achieve the decoupled dynamics of position and force controlled axes in the Cartesian space. Experimental studies on force tracking control performance are conducted. |
|
|
Experimental Studies on Neural Network Compensation for Time-delayed Control of a Robot Manipulator |
|
Seul Jung, Jun Woo Lee(Chungnam National University, Korea) |
| This paper presents the experimental studies of neural network control for a time-delayed control (TDC) method in robot manipulators. The accuracy of the inertia model of a robot manipulator is important for TDC implementation. Here the accurate estimation of an inertia model is improved by neural network to enhance the tracking control performance. Experimental studies of a robot manipulator to follow the circular trajectory are conducted to confirm the proposal. Tracking performances of circular trajectory tracking tasks by TDC and TDC with neural network are compared. |
|