This study presents a combination of soft computing techniques, namely backpropagation neural network, fuzzy and genetic algorithms that are used to control theBridgestone Hybrid Robot Arm (BHRA).The workspace of the BHRA?s end eï¬ector is divided into small segments and thetrajectory independent parameters of all these segments are learned by training smallsize (only three nodes) neural networks for each segment. The structure of these neuralnetworks is based on the physical model, which is derived from the Language-Eulermechanics of the robot arm. To maintain continuity on the small neural networks, weuse a basic fuzzy algorithm whose fuzzy membership function parameters are optimizedby genetic algorithm (GA). The proposed technique?s performance was compared withonly-neural network controller and shown to be more accurate in trajectory control forrubbertuator robots.The main goal of this study is to maintain a better oï¬-line control on the rubber-tuators by using only small size (3 nodes and one hidden layer) neural networks and asimple fuzzy algorithm with minimal linguistic variables and minimal number of rules.On the other hand, ï¬nding a better oï¬-line control ensures that small learning rateswill be suï¬cient for future on-line training control, and choosing small learning rateswill decrease the prospect of divergence and the risk of instability in control.