This paper presents a new approach to indirect vector control of induction motors. Two nonlinear controllers, one of sliding mode type and the other PI-fuzzy logic-based, define a new control structure. Both controllers are combined by means of an expert system based on Takagi-Sugeno fuzzy reasoning. The sliding-mode controller acts mainly in a transient state while the PI-like fuzzy controller acts in the steady state. The new structure embodies the advantages that both nonlinear controllers offer: sliding-mode controllers increasing system stability limits, and PI-like fuzzy logic based controllers reducing the chattering in permanent state. The scheme has been implemented and experimentally validated.

This paper describes the mechanical devices, the movements and the associated kinematic models of a novel wheelchair prototype capable of climbing staircases. The key feature of the mechanical design is the use of two decoupled mechanisms in each axle, one used to negotiate steps, and the other to position the axle with respect to the chair to accommodate the overall slope. This decoupling makes possible many different climbing strategies, the overall mechanism being extraordinarily versatile from a control point of view. Kinematic models have been developed for the different mechanical configurations that appear during all the ascend/descend processes. These models are required to control the actuators of the wheelchair in such a way that its centre of mass is able to follow arbitrary spatial trajectories. This is very important as one has to design very smooth spatial trajectories, keeping a near null inclination of the seat all the time in order to guarantee the comfort of the passenger, usually a handicapped or injured person. A real prototype is presented, and experimental results are reported that show the efficiency of the mechanism and the accuracy of the kinematic models developed.