Real-Time Optimisation of Control Strategy for a Low Fuel Consumption Vehicle Engine Using STM32 Microcontroller: Preliminary Results

The goal of this work is to analyse possibility of performing real-time optimisation of control strategy for a vehicle which leads to low fuel consumption using an embedded hardware platform with limited resources (the STM32 microcontroller). An original specialised car developed to take part in the Shell Eco-marathon competition is considered. At first, the optimisation task based on a vehicle model is discussed. In order to reduce on-line computational complexity, a relatively simple, but efficient parametrised control strategy is proposed. Next, since the majority of computation time is consumed by Ordinary Differential Equations (ODE) integration of the car model, a customised fast calculation method is developed using specific problem features. The discussed method is compared in terms of speed and memory usage with two classical approaches. It is approximately 10 times faster and is characterised by low memory demand.