2020 Fiscal Year Research-status Report
Research development of the super leading safe and smart next generation robot based on breakthrough superior wideband force perception and outstanding robot artificial intelligence
| Project/Area Number |
20K14713
|
|---|
| Research Institution | Nagaoka University of Technology |
|---|
Principal Investigator |
TRAN PHUONGTHAO 長岡技術科学大学, 工学研究科, 助教 (20848923)
|
|---|
| Project Period (FY) |
2020-04-01 – 2023-03-31
|
| Keywords | Motion control / Force control / Intelligent sensing / Intelligent control / Deep learning / Robot |
|---|
| Outline of Annual Research Achievements |
The ascending need of human-robot collaboration has led to the extensive evolution of human-robot interaction applications. To establish the effective human-robot collaborations, not only the successfully collaborative executions are required but the safe interaction is also critical to be realized. Therefore, the purpose of this research is to build the leading safe, smart and reliable human-robot collaboration system. The proposed methodology aims at achieving the key technology for force perception and intelligent force control to constitute the safe, smart and friendly robots for the future society.
To achieve this purpose, this research has been developing the breakthrough wideband force perception and outstanding intelligent robot force control. The breakthrough wideband force observer is proposed to achieve the fine force perception with the superior wideband force sensation which ensures the precise force perception, especially interaction force with high frequency range. The outstanding intelligent force control is realized by the intelligent force sensing and intelligent force controller to realize the safe and reliable interaction for robot system.
The superior wideband force perception of 1 kHz has been realized. The intelligent force sensing has been proposed to realize the high robustness to the variable noise characteristics, and to solve the problem of undesired and variable periodicities. The methodology of intelligent force controller has been being developed.
|
| Current Status of Research Progress |
Current Status of Research Progress
2: Research has progressed on the whole more than it was originally planned.
Reason
To achieve the safe and reliable interaction for robot system, the superior wideband force sensation ensuring the precise force perception is required. Therefore, in the first year, this research focused on the realization of superior wideband force sensation. Through experimental implementation, the force sensing bandwidth of the proposed force sensation method has been widened to 1 kHz. Normally, the human force perception bandwidth is 400 Hz. The bandwidth of 1 kHz is much wider than human tactile sensation, and is a record in the field of robot force control. The proposed outstanding fine force perception with this superior wide bandwidth ensures the proper force information for controlling interaction with human and environments.
Moreover, to achieve intelligent force control, the deep learning based intelligent force sensing to realize the high robustness to variable noise characteristics and variable periodicities has been proposed. The intelligent force sensation realization has been partially completed. Through simulations based on experimental testing, the force sensation achieved the fine force perception under the condition of variable noise characteristics, even in highly noise-contaminated conditions. This outcome and the successful implementation of deep learning based approach can foster the complete realizations of the intelligent force sensation and intelligent force controller in the next period of this research.
|
| Strategy for Future Research Activity |
To achieve the purpose of building the leading safe, smart and reliable human-robot collaboration system, this research continues to develop the remaining part of the intelligent force sensation and aims at realizing the complete intelligent force sensation with experimental demonstrations in the next year. Moreover, the superior wideband force sensation realized in the first year will be integrated with the intelligent force sensing to construct the outstanding fine force perception. In addition, the methodology of intelligent force controller will continue to be developed based on reinforcement deep learning.
|
| Causes of Carryover |
The research material intended to buy in the FY2020 could not be purchased because the releasing date has been delayed to the FY2021. Therefore, the amount intended to buy that research material will be transferred to the FY2021.
The remaining amount will be used to buy research material in the FY2021.
|
