A pneumatic slip control braking system was demonstrated, which reduces the emergency stopping distances of heavy goods vehicles by up to 19%. Solutions are still required to set the optimal reference wheel slip for this system online, so that it can adapt to changing operating conditions. This paper considers whether the use of extremum-seeking algorithms is a feasible alternative approach to online tyre model fitting, the computational expense of which has, to date, inhibited real-time implementation. The convergence and the stability properties of a first-order sliding-mode extremum-seeking algorithm are discussed, and its tuneable parameters are recast as physically meaningful performance metrics. Computer simulations are conducted using a detailed braking system model, and hardware-in-the-loop simulations are conducted with prototype pneumatic slip control braking hardware for heavy goods vehicles. The extremum-seeking algorithm enables the braking system to achieve at least 95% of the maximum possible braking force for almost the entirety of an emergency stop. The robustness to parameter errors, the road roughness and the changing friction conditions are all explored.
Extremum-seeking algorithms for the emergency braking of heavy goods vehicles
