6533b7d4fe1ef96bd12628c6

RESEARCH PRODUCT

Adaptive neuro-fuzzy inference system for kinematics solutions of redundant robots

Octavian BologaSever-gabriel RaczMihai CrenganisRadu-eugen Breaz

subject

Adaptive neuro fuzzy inference systemRobot kinematicsEngineeringInverse kinematicsbusiness.industryKinematicsRobot end effectorlaw.inventionRobot controlComputer Science::RoboticslawKinematics equationsControl theoryRobotbusinessComputingMethodologies_COMPUTERGRAPHICS

description

This written paper presents aspects concerning the implementation of the Adaptive Neuro-Fuzzy Inference System (ANFIS) in the resolution of a redundant serial robot kinematics. The kinematics solutions are divided into two categories: direct kinematics solutions and inverse kinematics solutions. To be able to control a robot the most important solutions are the ones for the inverse kinematics since one knows the position and the final orientation of the end effector and needs to determine the relative displacement or movements into the robot couplings. To obtain the optimal solutions for the inverse kinematics of a redundant robot the mathematical equations were based onto the redundancy circle method. The ANFIS model is used in order to determine the robot elbow position onto the redundancy circle so the robot will able to avoid different obstacles.

yearjournalcountryeditionlanguage
2016-05-012016 6th International Conference on Computers Communications and Control (ICCCC)
https://doi.org/10.1109/icccc.2016.7496773