Simulation design of trajectory planning robot manipulator

Wahyu S. Pambudi, Enggar Alfianto, Andy Rachman, Dian Puspita Hapsari

Abstract


Robots can be mathematically modeled with computer programs where the results can be displayed visually, so it can be used to determine the input, gain, attenuate and error parameters of the control system. In addition to the robot motion control system, to achieve the target points should need a research to get the best trajectory, so the movement of robots can be more efficient. One method that can be used to get the best path is the SOM (Self Organizing Maps) neural network. This research proposes the usage of SOM in combination with PID and Fuzzy-PD control for finding an optimal path between source and destination. SOM Neural network process is able to guide the robot manipulator through the target points. The results presented emphasize that a satisfactory trajectory tracking precision and stability could be achieved using SOM Neural networking combination with PID and Fuzzy-PD controller.The obtained average error to reach the target point when using Fuzzy-PD=2.225% and when using PID=1.965%.


Keywords


Neural network; Robot manipulator; Trajectory

Full Text:

PDF


DOI: https://doi.org/10.11591/eei.v8i1.1179

Refbacks

  • There are currently no refbacks.




Bulletin of EEI Stats