Perception of a robot’s environment is crucial to estimate its current state, localize and navigate its surroundings. Multiple strategies have been developed over the years to address this subject, in the name of visual servoing. A strategy this bachelor’s project thesis presents is a vision-based ‘tag and navigate system’ for multi-wheel autonomous and semi-autonomous robot systems leveraging the tools of monocular visual servoing. This project aims to develop a vision-based electronic leash(hence the name VBEL) for multi-wheel robot systems, a computationally cost-effective solution . The decision made by the object tracking algorithm in tandem with other sensor fused feedback (IR and Ultrasonic) will tag the robot to the object of interest and track it actively. The visual input captured by the camera is processed to arrive at a characteristic feature set, world coordinates, and robot’s estimated state used to further arrive at a decision that will track the target object - a person, a ball, or a number plate on a vehicle - and navigate towards it.