Self-guidance Control Of A Multi-articulated Vehicle

Bus systems in major cities often have to deal with constantly growing demand for higher capacity and reliability.Traffic jams caused by private transportation as well as capacity constraints due to the limited number of seats in the vehicles are complicating factors in this situation.Railway system can easily be adjusted to meet the required number of passengers to be transported while being independent from private traffic.In contrast to this,establishing a railway system depicts an expensive task in terms of money,time and the amount of constructional work.An alternative solution to meet the requirements without having to lay train tracks emerged from the idea to establish a BusRapid-Transit(BRT)system which is oriented towards train systems in terms of capacity and reliability.To meet the requirements new vehicle structures with more than one hinge were developed in the past.Tendencies point towards multi-articulated vehicles that are extendable by an arbitrary number of modules.Current research deals with satisfying maneuverability requirements that emerge e.g.from trace curves which occur when trying to steer a multi-articulated vehicle only using the steering angle of the front axle.Equipped with all wheel steering functionality the same requirements like conventional single-articulated busses are supposed to be met.In this paper a self-guidance control algorithm is equipped to a multi-articulated vehicle with N-bodies.It is supposed to minimize off-tracking and turning radii to guarantee maneuverability and pedestrian safety in urban scenarios.The first step is to work out a proper way to control the steering using the hub motors attached to each wheel.The main concern of this paper is a performance examination of the steering strategy in different scenarios.Hereby a PI-control based on a sufficient kinematic model is supposed to be verified while the tracked path corresponds to realistic curvature constraints.The second step is to consider a more realistic vehicle model using a dynamic model.