BACK-DRIVING A TRUCK WITH SUBOPTIMAL DISTANCE TRAJECTORIES - A FUZZY-LOGIC CONTROL APPROACH

This paper provides a suboptimal solution to the problem of automatica lly bpack-driving a truck using the natural parabolic paths as the sho rtest moving distance requirement. By applying fuzzy logic control tec hniques, a controller with nine rules is developed, which works well e ven without using a mathematical system model, As long as the states ( position and orientation) of the truck are measurable at each discrete -time step during the control process, this controller can drive the t ruck to follow any feasible trajectories (their smallest radii are no less than the smallest radius of the curve along which the truck can t ravel), and to move successfully into a prescribed parking lot, In add ition to the design of the controller, controllability and stability o f the control system are briefly discussed under the condition that on ly partial information about the current states of the system are avai lable, Simulation results are presented, to demonstrate the accuracy a nd effectiveness of this new fuzzy logic controller and to compare its control performance with other fuzzy logic controllers that were desi gned for the same purpose under the same conditions but without using any optimality criterion.