Abstract

Brake-by-wire (BBW) is a fundamental function required by intelligent vehicles. And the One-Box electrohydraulic BBW (EHB) system is becoming the mainstream BBW solution. To improve the wheel pressure regulation (WPR) ability in the One-Box solution, this study proposes a new WPR scheme by directly coordinating the adjusting valves and the motor-powered booster instead of the traditional plunger pump. First, a disturbance rejection adaptive control (DRAC) method is proposed for the pressure control in the master cylinder during this special WPR process, which can deal with parameter uncertainty and disturbance simultaneously to maintain stable backpressure. Then, a novel adaptive neural network controller (ANNC) is constructed for the control of adjusting valves. The proposed ANNC consists of two radial basis function neural networks (RBFNNs) that can learn the system dynamics and open-loop control characteristics in real-time and require few computational resources. Finally, hardware-in-the-loop (HIL) experiments are conducted, and the results proved the superiority of the booster-based WPR scheme with the proposed control methods (DRAC + ANNC) compared with the method under the traditional pump-based WPR scheme by 19.8% and 31.6% in Multistep and Sine pressure tracking scenarios, respectively. This will further enhance the precise braking ability of BBW vehicles.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call