The objective of this paper is to propose a method for navigating a vision-based service robot using stereo vision and Kalman Filtering. We have develop modules for face recognition system, visual tracking and multiple moving obstacles avoidance using stereo camera. Someone that walks will be identified as moving obstacle and we propose a method to avoid the multiple moving obstacles. Kalman filtering for distance estimation using stereo camera used for stability of measurement and removing noise from the input images. To overcome the inaccuracies of vision sensor, Bayesian approach is used for estimate the absense and direction of obstacles. Algorithms for moving obstacles avoidance method proposed and experiment results implemented to a service robot also presented. Our system can overcome the drawbacks of popular obstacle avoidance methods because it provides a complete mechanism for a vision-based service robot so that it can avoid the moving obstacles.Various experiments show that our proposed method very fast, and successfully implemented to service robot called Srikandi III in our laboratory.