|
|||||
|
|
Learning how to grasp based on neural network retraining |
|
| Authors: | Leonardo M Pedro Valdinei L Belini Glauco AP Caurin |
| Affiliation: | 1. Center for Exact Sciences and Technology, Federal University of S?o Carlos , S?o Paulo , Brazil lmpedro@ufscar.br;3. Center for Exact Sciences and Technology, Federal University of S?o Carlos , S?o Paulo , Brazil;4. Engineering School of S?o Carlos , University of S?o Paulo , S?o Paulo , Brazil |
| Abstract: | Humans have an incredible capacity to manipulate objects using dextrous hands. A large number of studies indicate that robot learning by demonstration is a promising strategy to improve robotic manipulation and grasping performance. Concerning this subject we can ask: How does a robot learn how to grasp? This work presents a method that allows a robot to learn new grasps. The method is based on neural network retraining. With this approach we aim to enable a robot to learn new grasps through a supervisor. The proposed method can be applied for 2D and 3D cases. Extensive object databases were generated to evaluate the method performance in both 2D and 3D cases. A total of 8100 abstract shapes were generated for 2D cases and 11700 abstract shapes for 3D cases. Simulation results with a computational supervisor show that a robotic system can learn new grasps and improve its performance through the proposed HRH (Hopfield-RBF-Hopfield) grasp learning approach. |
| Keywords: | robotic grasping neural network retraining mesh simplification Hopfield neural network robot learning |