| Full text | |
| Author(s): |
Lahr, Gustavo J. G.
;
Garcia, Henrique B.
;
Silva, Thiago H. S.
;
Caurin, Glauco A. P.
;
Fleury, AD
;
Rade, DA
;
Kurka, PRG
Total Authors: 7
|
| Document type: | Journal article |
| Source: | PROCEEDINGS OF DINAME 2017; v. N/A, p. 13-pg., 2019-01-01. |
| Abstract | |
This paper presents the implementation of an optimization method to find, without knowledge of the environment characteristics, the best interaction controller parameters to revamp the coupled dynamics. The objective is to improve various industrial robot applications that involves mechanical contact. An enhanced contact is accomplished by lowering the following metrics: rise time, total variation and steady state error. Hence, the impedance controller was the interaction control technique chosen to be optimized. Contact is established between a Kuka KR16 robot TCP and an aluminum platform, where the force data was acquired by a 6-axis force-torque sensor located in the robot's end-effector. Using a hardware-in-the-loop optimization approach, the force feedback is processed by a NSGA-II algorithm. Each individual of the GA represents a specific impedance controller and as the generations passes, these values get more suitable for lowering the metrics. Results show convergence in 5 generations. (AU) | |
| FAPESP's process: | 15/07484-9 - Real-time movement synthesis for dynamic robot precision assembly task |
| Grantee: | Gustavo Jose Giardini Lahr |
| Support Opportunities: | Scholarships in Brazil - Master |
| FAPESP's process: | 15/24343-0 - Comparative analysis between machine learning reinforcement methods applied in aeronautics furniture production |
| Grantee: | João Victor Rocha Soares |
| Support Opportunities: | Scholarships in Brazil - Scientific Initiation |
| FAPESP's process: | 13/07276-1 - CEPOF - Optics and Photonic Research Center |
| Grantee: | Vanderlei Salvador Bagnato |
| Support Opportunities: | Research Grants - Research, Innovation and Dissemination Centers - RIDC |