Artificial neural networks controller of active suspension for ambulance based on ISO standards

Abstract

The level of vibration to which a patient is subjected in an ambulance is often too high. Ambulance personnel should take steps to reduce these vibrations. To avoid slowdowns and other barriers that create high vibration peaks, the speed is reduced or the ambulance deviates from the quicker or shorter path. This work implements Artificial Neural Network (ANN) control over five low-cost active shock absorbers proposed to decreasing the impact of vibration on a patient’s body during an ambulance ride. For this, the passive shock absorber is replaced by a new actuator consisting of a conventional hydraulic cylinder with a proportional butterfly valve placed outside the cylinder between its orifices. The ANN is used to adjust the damping coefficient. The ANN controller inputs are the accelerations of the sprung and unsprung masses, and the output is the valve opening area. Two forces may be exerted by this active suspension system: one is used to compensate for the mass of the stretcher, the patient and the medical equipment if present and a second is used to actively isolate the patient from the vibrations of the ambulance. The performance of the active suspension ambulance is contrasted to that of a traditional ambulance, in which the stretcher is rigidly linked to the ambulance body. When compared to alternative controllers, the findings showed that the proposed register controlled by ANN performed better. The simulations demonstrate that the active system may minimize the vibrations of the patient’s mass and the stretcher by more than 70% for random abnormalities in the road that meet the ISO2631-5 and ISO 8608 standards.