| Presentation | 2014-12-13 Forward-Propagation Learning Rule to Acquire the Inverse Dynamics Model of an Arm with Coulomb's Friction Yu KIYOSAWA, Takahiro KAGAWA, Yoji UNO, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | A forward-propagation learning rule has been used to acquire inverse models of controlled objects in a neural network as a feedforward controller. However, an approximate inverse model must be acquired before learning. We have proposed to combine a feedback controller with the forward-propagation learning rule. From randomized initial weights, the realized trajectory can converge to the desired trajectory by a few iterations of this learning scheme. In this report, our learning scheme was applied to acquire inverse dynamics models of a two-link arm with coulomb's friction and the efficacy was confirmed by computer simulation. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Forward-propagation learning rule / inverse model / neural network / feedback controller / coulomb's friction |
| Paper # | NC2014-49 |
| Date of Issue |
| Conference Information | |
| Committee | NC |
|---|---|
| Conference Date | 2014/12/6(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | |
| Vice Chair | |
| Secretary | |
| Assistant | |
| Paper Information | |
| Registration To | Neurocomputing (NC) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Forward-Propagation Learning Rule to Acquire the Inverse Dynamics Model of an Arm with Coulomb's Friction |
| Sub Title (in English) | |
| Keyword(1) | Forward-propagation learning rule |
| Keyword(2) | inverse model |
| Keyword(3) | neural network |
| Keyword(4) | feedback controller |
| Keyword(5) | coulomb's friction |
| 1st Author's Name | Yu KIYOSAWA |
| 1st Author's Affiliation | Graduate School and School of Engineering, Nagoya University() |
| 2nd Author's Name | Takahiro KAGAWA |
| 2nd Author's Affiliation | Graduate School and School of Engineering, Nagoya University |
| 3rd Author's Name | Yoji UNO |
| 3rd Author's Affiliation | Graduate School and School of Engineering, Nagoya University |
| Date | 2014-12-13 |
| Paper # | NC2014-49 |
| Volume (vol) | vol.114 |
| Number (no) | 362 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |