Motion planning for intelligent cars following roads based on feasible neighborhood

A motion planning method for intelligent cars following roads based on feasible neighborhood is presented, with which we can break down a dynamic and complex control process into a series of static and simple ones. The method can imitate the eyeshot of a skilled driver driving a car on road, and may be an available approach for the control of automatic driving. According to a car's orientation relative to road, a feasible neighborhood system consisted of trapezoids for intelligent cars are presented. An experimental approach to find satisfactory feasible neighborhoods for autonomous vehicles following roads are discussed, which are built by considering the changes of road edges. By the theories of neighborhood control systems, the whole motion planning for intelligent vehicles following roads is no other than the motion planning in finite simple feasible neighborhoods. Each of the motion planning in finite local feasible neighborhoods can be easily realized by fuzzy control rules and obtain a satisfactory result. All the methods are designed basing on the original manners of data from available sensors of angle and distance. So do the simulations on the control processes. The effectiveness of the presented methods is demonstrated by several simulation results with a full size car on road under simple fuzzy control.