The e-puck TM mobile robot is used and an intelligent obstacle avoidance algorithm is developed in this paper. The image data are processed by edge detection method. By using the recurrent fuzzy neural network (RFNN), the horizontal edge (HE) and vertical edge (VE) are feed into RFNN to train the control rules such as to control the right and left wheels of e-puck robot to avoid obstacles. The good control performances and effectiveness are demonstrated by the simulations of Matlab TM and Webots TM ; meanwhile, the empirical tests are also implemented to verify these performances. (Chih-Min Lin). board of e-puck robot is based on a 32-bit microprocessor and provides wireless blue-tooth network support. The microprocessor extension board runs in parallel with the microprocessor on the e-puck motherboard with communication between the two via communication bus. The extension board is designed to handle computationally intensive image processing, wireless communication and high-level intelligent robot control algorithms, data processing and motor control [2].In this paper, the Webots TM software is used, the dynamic image from VGA camera characteristics of the e-puck robot model can be exploited for investigating the temporal aspects of multimodal integration. The temporal window for integration is shown to have an impact on the multisensory interaction, so the possibilities for its adaptation within the control field model and its impact on the computational outcomes are easily investigated [3]. Webots TM running on Windows is intended for researchers and teachers interested in mobile robotics. It is commercially available from Cyberbotics Ltd. Although the final aim is real robotics,