Mobile Robot Path Tracking Control

The path tracking control algorithm of wheeled mobile robot, which determines robot's thinking, decision-making and behavior, is studied here by applying the Generalized Predictive Control algorithm (abbreviated as GPC), smith forecast theories and filter theories. The control module has been successfully designed and been proven to be efficient by some experiments.Firstly, the mathematic and physical models of mobile robot have been set up according to the well-known environment. Due to the nonholonomic constraint, the mobile robot system belongs to nonlinear system. The corresponding linear system can be obtained by deploying the system at work point using Taylor equation. In this case, the high-level item is omitted after Taylor deployment, and then GPC algorithm is applied. GPC algorithm which has few model parameters and keeps the merit of the self-adaptive control algorithm can be divided into multi-step prediction, rolling optimization and feedback verification. Smith forecast control arithmetic has made the hysteresis phenomena to be away from the feedback loop which improves the system's response performance. Simulations indicate that the system's robust and anti-jamming performances have been greatly improved.Secondly, the GPC controller has been used in the path tracking and orientation of our wheeled mobile robot, which has a good performance in the 'ABU University Robot Domestic Contest 2006'. The software and hardware of the control system have been all-sided considered and carefully designed. The response performance of control system has been improved by the feedback loop which consists of coders, color sensor, and GPC controller.Finally, applying the programmes of position detecting and compensative control, some experiments have been carried out to validate the efficiency of the control algorithm when the preset paths are line, circle and curve. Analysis of these experiment data show that wheeled mobile robot with this control algorithm can accurately follow desired paths and eliminate the influence of disturbances.