Simulation of robot path planning by 3-dimensional (3D) visualization using neuro fuzzy systems (NFS)

A great deal of interest has been shown in the literature in developing different clustering techniques. The main objective of clustering is to select a reduced data set from an original data set so that the subsequent portion of the problem can be handled by taking less computational time. The purpose of the paper is to show that sometimes the clustering techniques do not take into account the critical object points which are to be included in the subsequent portion of problem. A 3D visualization of a robot path to avoid an obstacle has been simulated. By using only clustering techniques, the mobile robot in the simulation of the virtual lab is unable to avoid the obstacle. However, when some critical points are forced into the clustered data, the performance of avoiding an obstacle is relatively improved. The obstacle avoidance problem can be handled in a computationally intensive way. A procedure to handle this problem is discussed. The modified clustering technique may be able to draw better results to implement this problem.