Community Research and Development Information Service - CORDIS


This paper reviews the use of neural network techniques for achieving adaptability in both manipulator and mobile robots. First, the different learning approaches are classified according to the amount of training information they require: quantitative (supervised approaches), qualitative (reinforcement-based approaches) or none (unsupervised approaches). Afterwards, the adequacy of each approach for solving specific problems in robot control is illustrated through four working industrial prototypes. The problems tackled are the inverse kinematics and inverse dynamics of robot manipulators, visual robot positioning and mobile robot navigation.

Additional information

Authors: TORRAS C, Institut de Cibernètica (CSIC-UPC), Barcelona (ES);CEMBRANO G, Institut de Cibernètica (CSIC-UPC), Barcelona (ES);WELLS G, Institut de Cibernètica (CSIC-UPC), Barcelona (ES);MILLAN J DEL R, JRC Ispra (IT)
Bibliographic Reference: Paper presented: International Workshop on Artificial Neural Networks, Malaga (ES), June 7-9, 1995
Availability: Available from (1) as Paper EN 39079 ORA
Record Number: 199511092 / Last updated on: 1995-08-18
Original language: en
Available languages: en