ABSTRACT
This chapter presents the visual trajectory tracking tasks for wheeled mobile robots including both eye-to-hand and eye-in-hand configurations. Due to the nonholonomic constraints of wheeled mobile robots, the error system development, control design, and stability analysis in this effort requires fundamentally new control approaches in comparison to general robotic systems, a typical feedback linearizing controller developed for holonomic systems cannot be used to solve the problem. For the eye-to-hand case, an adaptive visual servo controller is developed for the off-board, monocular camera configuration that ensures global tracking for a mobile robot. The wheeled mobile robot(WMR) is controlled by a torque input applied to the drive and steer motors. Vision-based control methods for WMR are proposed for camera-to-hand and camera-in-hand configurations. The camera-to-hand case, with the visual feedback from a fixed monocular camera, the position/orientation of a WMR is controlled to follow a desired time-varying trajectory which is defined by a pre-recorded image sequence.
